"We congratulate the LibreOffice community on their success over their inaugural year and wish them luck in their future endeavors. We look forward to opening up the dialogue between Open Document Format-oriented communities to deepen understanding and cease the unwarranted spread of misinformation."
Interesting. I wonder how exactly they see their relationship with libreoffice going forward.
Regardless, this is great news. Apache cannot possibly mishandle the project more than Sun/Oracle were.
I would completely understand that the LibreOffice folks are bitter about the whole Sun/Oracle situation, but you'd think it would be in the best interest of everyone to reconcile and merge the two branches. Yes, Apache Foundation would probably add bureaucracy, but that's not always bad.
I don't see the point really. LibreOffice already has working infrastructure, did loads of cleanup and has a working community. I don't see the point of the Apache Foundation, nor rewriting stuff just so it can have a different license.
Which community is bigger? Which app suite has a larger install base. The problem is that neither app is quite comparable to Office, in my opinion; neither has stolen market share from Office. A divided community will close the gap even slower.
I was commenting on the remark that the fork should be merged.
You cannot modify the source of Microsoft Office. Comparing office suites is an entirely different topic.
I don't see a divided community. One project is really active and alive, then a year later OOorg tries to start something else.
It seems inefficient to have this split. But that is the same for everything. Why have different languages, multiple OSes, multiple distributions, desktop environments, etc. Might be inefficient, but unless you take away peoples freedom to do what they want, I don't see this changing.
Trying to compete with Microsoft Office seems futile anyway. I'd rather have a focus on making a good office suite. It doesn't really matter if it can do everything that Microsoft Office can.
I know that is the reason for starting the OOorg on Apache, so the license stuff is easier for IBM.
I just don't think it'll work. Most distribution companies prefer GPL because in that case everybody benefits. You cannot take the work of all companies, then do your on thing on top and don't give back.
Believe it would be easier to not have Lotus Symphony open source. That, or change strategy and rely on brandname. I don't think it matters much that something is fully open sourced and can be reused by another company.
Marketing, trademark and strong brand (meaning: trusted, supported, reliable) goes a long way. Can still ask money for it, and if someone wants to give it away for free they'll have to use a different name for it (similar to the Firefox name in Debian).
The Iceweasel name in Debian is not because Debian give away Firefox for free (as in no cost), it's because Mozilla only allow you to call the software Firefox if you use an official build (or get permission for non-official builds), and I believe the artwork also has a proprietary licence, making it non-free by Debian's definition.
0x12: Please explain "the only thing". I don't see this; it way more complex than this. To me it seems that the different options+projects is why we have really good solutions, even if sometimes some things take a bit longer to be developed.
e.g. project a is developed. Design is wrong according to another group, they develop project b, then maybe you get project c. Messy on the short term. Long term, seems only the high quality projects survive.
I don't see "only the high quality projects" surviving in the long term. Using and watching Linux since 1997, I mostly see a constant, perpetual flux and incohesiveness. And a lot of damn fine projects abandoned.
For example, Nautilus (the file manager, in it's original design). ReiserFS. Omega (the unicode alternative TeX implementation). The Fresco windowing system. Enlightenment E17. The Parrot VM. ...
Actually it was. Current day Nautilus has little to do with the original scope and functionality of the project as envisioned (and promoted) by Eazel. It's like saying Hurd still exists. Yeah, but hardly anybody uses it and it failed to fulfill what it promised for 25 years.
> IIRC E17 is used on various phones
Yeah, which all of 5 people use, and one of them is Rasterman.
> Reiser killed someone
Isn't an often repeated premise of OSS that a project can keep on despite being abandoned by its' leaders?
> Parrot is still under development
Still in perpetual flux.
> Perpetual flux and incohesiveness is too management speak for me.
It's also something felt everyday by common people and expert hard core users alike.
> Look inside a big company, pretty much the same thing going on.
Companies tend not to fuck too much with the flagship programs, revenue streams, and seldom just for the "fun of it".
How many times, for example, has the KDE multimedia infrastructure been replaced with something entirely different only for this to happen again?
LibreOffice is now default Office suite in OpenSUSE, Fedora, Ubuntu, and maybe others. So I guess that that the LibreOffice will get a larger user base soon.
That was done as a reaction to Oracle, if OpenOffice becomes viable again and gains traction it can change back.
LibreOffice's staying power depends on how much of their effort was simply "sticking it to Oracle", if they get bored and fade away then so will LibreOffice.
Apart from nerds there is no loyalty to the product just to getting the job done, if the word processor opens their documents then users simply don't care. Similarly most users don't use Office because it is Office, they use it because it is there.
Whichever way they want to go about it, they should just merge already. Yes, they are different options in the same space with different ideas, but now is when the codebase is the closest it will likely ever be to being the same.
The problem really is the license. LibreOffice is GPL, OpenOffice will be the Apache License, the difference being GPL requires the source of modifications to be released as GPL as well. There are many people who would want to contribute to something that isn't GPL, especially companies. A consequence of the licensing means LibreOffice can merge changes from OpenOffice, but OpenOffice can't do the same.
GPL can be considered too free, forcing to to release your source code when you don't want to. This may not be desirable for some people, especially some companies, so they prefer non-GPL software.
A real application of this is using a GPL piece of software in a much larger project (like a library). This would mean you would have to release the entire project as GPL, even if you don't want to release other parts as GPL. This is where things like LGPL and dual-licensing can come in, allowing you the 'freedom' to release your code with the license you want.
If you contribute to an open-source project, your code is already released. What GPL prevents is a competitor using your code released as open-source and turning it into a competing proprietary product. Apache licenses allow that. if you fear your competitors may use your code, contributing to an Apache-licensed project is infinitely worse because then your competitors are allowed to derive their work from your contributions and never share back, something GPL would prevent.
It also prevents YOU from extending the product in some way, embedding it in your proprietary stuff, and then keeping the extensions you wrote secret from the rest of the world. Now, ME, I wouldn't be interested in keeping my extensions secret, because I don't really have anything to hide. But some folks, especially big companies, DO have something to hide. Perhaps, (not a real example) they mean to release a tablet with office software built in, but with a back-door built into the office software that allows it to secretly track all documents created and report back to the mother-ship. At least I HOPE that's not a real example.
Actually that is not true. If you don't give someone the binary, you don't have to give the source.
So you can, if you want, sell the binary to select people (and ask them nicely not to give it to anyone else), and you only have to give the source to those select people, but not to anyone else.
The GPL does not require public source, only that the source follows the binary.
Or you can turn your product into a web application which you host, and ignore the GPL entirely if it's not the Affero version. The same might apply if you have a small client binary which you distribute the code to, but keep the server code (based on a GPL project) secret.
LibreOffice is not GPL, it is LGPL (and also MPL for new contributions, which I believe is not GPL compatible). So depending on how OpenOffice.org used parts of LibreOffice, they might be able to include them. Admittedly it starts to get complicated when you have three licences involved.
I'm not sure what the benefit would be to LibreOffice if they merged under the ASF. It's already a large project with lots of contributors (and an excellent 'how to get involved' section which more OSS projects should emulate), infrastructure, mailing lists etc., so what would be the benefit of being an Apache project?
In my opinion, the best solution would be to leave OpenOffice.org to bitrot and ask developers to contribute to LibreOffice, given that it is the preferred suite of nearly every distribution already.
I doubt they'll be much competition for libreoffice. Given Apache's repertoire I wouldn't be surprised if they went the cloud/web route and tried to compete with google docs or office 365.
The biggest problem for OpenOffice.org as an Apache project, especially given the existence of a vibrant LibreOffice, is what companies are going to fund OpenOffice.org development and why? IBM will because they have a commercial product, sure, but who else? If there are no other companies funding significant development efforts (which in practice means they have to have a commercial product), then the rationale for a company to do all of their development in the open (and possibly risk a competitor taking their code and using it in a competing product with little or no effort contributed) tends to destroy the value of the Apache project.
With Libreoffice taking away a large number of the community developers, it's not clear there will be enough non-IBM corporately funded developers to make OpenOffice.org successful.
I'm not sure that rumor is correct. Look at this comment to Brian Proffitt's article, "Has IBM kicked OpenOffice.org to the curb?". It's from an IBM'er claiming that IBM is investing a huge amount into the OpenOffice project:
This whole thing was a fuck-you move by Oracle and I am surprised that Apache would be a part of this. They should defer to Libre Office and donate everything to them.
Can you expand on this? How is releasing their code under the Apache license, a less restrictive license than the libreoffice project uses, a "fuck-you move" to anyone?
Because they could have simply donated everything to Libre Office itself, instead of donating it to another organization under another license. There was no reason to do that aside from being sore losers.
Can you give a coherent explanation why the Republicans and Democrats should not merge?
EDIT: Thank you, downvoters. I look forward to your excellent answers that do not involve politics and ideology. I'm sure you just downvoted me while you began working on your much better answers to the parent's question and I look forward to reading them.
The LibreOffice project hoped they would, but Oracle didn't like the idea...
Probably the biggest thing is that IBM wanted it with a more permissive licence so they could keep selling their commercial product without giving away any of their source, but the LibreOffice project wanted to keep the suite as free software (which is why all the new code that LibreOffice writes is LGPL).
As it is, OpenOffice at Apache will probably bitrot - pretty much all Linux distros have switched to LibreOffice (so that's where all the developers who aren't employed by IBM or Oracle have gone), and LibreOffice can use Apache code in their product, but Apache OpenOffice can't use LibreOffice's code - so they won't benefit from the massive cleanup that the LibreOffice project has done to the codebase over the last year, or the new features they have brought in.
Unfortunately this doesn't matter given how far Libre Office has come since splitting off. I mean Libre Office already stated they have mobile version in the works, not to mention the other improvements they've brought to the project causing a lot of Linux distributions to switch to Libre Office as there default word processing suite.
A little too late for Oracle and the Apache Foundation in my opinion.
49 comments
[ 4.0 ms ] story [ 130 ms ] threadInteresting. I wonder how exactly they see their relationship with libreoffice going forward.
Regardless, this is great news. Apache cannot possibly mishandle the project more than Sun/Oracle were.
You cannot modify the source of Microsoft Office. Comparing office suites is an entirely different topic.
I don't see a divided community. One project is really active and alive, then a year later OOorg tries to start something else.
It seems inefficient to have this split. But that is the same for everything. Why have different languages, multiple OSes, multiple distributions, desktop environments, etc. Might be inefficient, but unless you take away peoples freedom to do what they want, I don't see this changing.
Trying to compete with Microsoft Office seems futile anyway. I'd rather have a focus on making a good office suite. It doesn't really matter if it can do everything that Microsoft Office can.
I just don't think it'll work. Most distribution companies prefer GPL because in that case everybody benefits. You cannot take the work of all companies, then do your on thing on top and don't give back.
Believe it would be easier to not have Lotus Symphony open source. That, or change strategy and rely on brandname. I don't think it matters much that something is fully open sourced and can be reused by another company.
Marketing, trademark and strong brand (meaning: trusted, supported, reliable) goes a long way. Can still ask money for it, and if someone wants to give it away for free they'll have to use a different name for it (similar to the Firefox name in Debian).
e.g. project a is developed. Design is wrong according to another group, they develop project b, then maybe you get project c. Messy on the short term. Long term, seems only the high quality projects survive.
For example, Nautilus (the file manager, in it's original design). ReiserFS. Omega (the unicode alternative TeX implementation). The Fresco windowing system. Enlightenment E17. The Parrot VM. ...
Perpetual flux and incohesiveness is too management speak for me. Loads of people who scratch an itch working out in the open.
Look inside a big company, pretty much the same thing going on.
Actually it was. Current day Nautilus has little to do with the original scope and functionality of the project as envisioned (and promoted) by Eazel. It's like saying Hurd still exists. Yeah, but hardly anybody uses it and it failed to fulfill what it promised for 25 years.
> IIRC E17 is used on various phones
Yeah, which all of 5 people use, and one of them is Rasterman.
> Reiser killed someone
Isn't an often repeated premise of OSS that a project can keep on despite being abandoned by its' leaders?
> Parrot is still under development
Still in perpetual flux.
> Perpetual flux and incohesiveness is too management speak for me.
It's also something felt everyday by common people and expert hard core users alike.
> Look inside a big company, pretty much the same thing going on.
Companies tend not to fuck too much with the flagship programs, revenue streams, and seldom just for the "fun of it".
How many times, for example, has the KDE multimedia infrastructure been replaced with something entirely different only for this to happen again?
LibreOffice's staying power depends on how much of their effort was simply "sticking it to Oracle", if they get bored and fade away then so will LibreOffice.
Apart from nerds there is no loyalty to the product just to getting the job done, if the word processor opens their documents then users simply don't care. Similarly most users don't use Office because it is Office, they use it because it is there.
A real application of this is using a GPL piece of software in a much larger project (like a library). This would mean you would have to release the entire project as GPL, even if you don't want to release other parts as GPL. This is where things like LGPL and dual-licensing can come in, allowing you the 'freedom' to release your code with the license you want.
So you can, if you want, sell the binary to select people (and ask them nicely not to give it to anyone else), and you only have to give the source to those select people, but not to anyone else.
The GPL does not require public source, only that the source follows the binary.
In my opinion, the best solution would be to leave OpenOffice.org to bitrot and ask developers to contribute to LibreOffice, given that it is the preferred suite of nearly every distribution already.
With Libreoffice taking away a large number of the community developers, it's not clear there will be enough non-IBM corporately funded developers to make OpenOffice.org successful.
http://www.itworld.com/it-managementstrategy/212949/has-ibm-...
OpenOffice.org was a pretty shitty trademark to begin with.
EDIT: Thank you, downvoters. I look forward to your excellent answers that do not involve politics and ideology. I'm sure you just downvoted me while you began working on your much better answers to the parent's question and I look forward to reading them.
I downvoted you for not adhering to your own standards.
Strangely, choosing Democrats and Republicans to illustrate that point was probably a poor choice, considering.
Probably the biggest thing is that IBM wanted it with a more permissive licence so they could keep selling their commercial product without giving away any of their source, but the LibreOffice project wanted to keep the suite as free software (which is why all the new code that LibreOffice writes is LGPL).
As it is, OpenOffice at Apache will probably bitrot - pretty much all Linux distros have switched to LibreOffice (so that's where all the developers who aren't employed by IBM or Oracle have gone), and LibreOffice can use Apache code in their product, but Apache OpenOffice can't use LibreOffice's code - so they won't benefit from the massive cleanup that the LibreOffice project has done to the codebase over the last year, or the new features they have brought in.
A little too late for Oracle and the Apache Foundation in my opinion.