Well, my prediction[1] came true even faster than I anticipated. I don't understand why so many people seem so desperate to see AOO fail and go away. Hey look, if you prefer LO, fine. Some of us prefer AOO for our own reasons. There's no reason to take glee in seeing a project struggle.
Anyway, none of this means the project is actually going to be retired. It's just a discussion around something that might happen.
> I don't understand why so many people seem so desperate to see AOO fail and go away.
In my case, the reason is splash-damage from the inverse-goodwill I have for Oracle and how their stewardship of OO (or lack thereof). To start off Sun handled the OOO-patchset (proto-LibreOffice) poorly, they should have merged it to mainline. Then Oracle came along and shat on the community leading to an outright fork. This I guess is par for the course for Oracle (see Hudson/Jenkins for another Oracle-instigated implosion).
I think the Apache Software Foundation allowed themselves to be 'used' by allowing Oracle to dump a dying AOO into their hands. At the time, it was clear that LibreOffice had won the the war that started in the OOO-patchset days, I don't know what the ASF thought would happen, but I hope they learn something from it.
At the time, it was not clear what the outcome of the community split was going to be. But what tipped the scales was getting the assets out of Oracle.
Even if the nascent project had only managed to receive the code and make one release before folding, that was worthwhile enough to justify accepting the donation. The project managed to flourish for some time after that, so in that regard it was well above baseline.
I think the Apache Software Foundation allowed themselves to be 'used' by allowing Oracle to dump a dying AOO into their hands.
"Used" how? Apache taking on AOO has been a win for Open Source in the general sense regardless of what happens to the project going forward. A large base of code which was previously locked up under Oracle's copyright is now licensed under the ALv2 for perpetuity.
At the time, it was clear that LibreOffice had won the the war that started in the OOO-patchset days,
There's no "war". Maybe the LO people see it that way, but we don't. AOO is for people who want AOO. The people who prefer LO (or something else) are not our enemies. They're just people with different preferences.
I don't know what the ASF thought would happen, but I hope they learn something from it.
What did happen? IBM donated a bunch of code, volunteers have contributed more, and AOO is a better product today than what it was when Oracle handed over the code. That's a Good Thing. None of which is to try and compare AOO to LO, and none of which is to say that AOO doesn't have flaws. But it's a product that has been used by a ton of people for productive ends over the past few years.
I will reitatate that I was explaining why I have ill-will towards Oracles handling of the situation. I am not affiliated with LO beyond being a happy user, I have never been a maintainer or a regular contributor (to any project for that matter).
> "Used" how? Apache taking on AOO has been a win for Open Source in the general sense regardless of what happens to the project going forward.
A better win for Open Source in general (IMO) would have been Oracle handing over the IP to the Document Foundation- that's were the OO developers and community were. I have nothing againts the ASF, and nothing but contempt for Oracle's behavior.
From where I stand, Oracle 'used' the ASF donation to spite the DF - you may agree or disagree on that point.
> There's no "war
It was a figure of speech.
> What did happen?
Let's recap: the ASF accepted the donation of a project that had lost developers to a competing fork, and was showing glaring signs of having lost steam. More importantly, the volunteers/'community' had moved on to LO[1]. Now the ASF is finally contemplating the idea of putting down the project. I would be sad if the ASF concludes that there is nothing to be learnt from this episode.
1. A single datapoint: Virtually all Linux distros shipped LO
What reasons are there to prefer OpenOffice? From my perspective, I hear about LibreOffice updates often, and OpenOffice doesn't sound like it's being maintained as much. I don't know if this is actually correct, but I never understood what the need was for keeping them separate after Oracle gave up OO to Apache. LibreOffice seems to have 'won' popular attention if nothing else, and is the one included in distros now.
Having two forks competing just seems confusing and a waste of effort on maintaining two forks. Initially it made sense to have a fully open fork when OpenOffice was under Oracle's control, but I don't understand what the purpose the two forks serve now that OO is under Apache's control.
The two forks represent two different ideologies to a large extent - "permissive" licensing vs "copyleft". Some of us think it's important to have a full-featured, world-class office suite that's under a permissive license. Others think it's important to have a full-featured, world class office suite that's copyleft. Who's to say either is wrong?
Feature wise, there may not be a lot of reasons to prefer AOO right now. I don't know, as I spend very little time looking at / using, LibreOffice. AOO does everything I need, so what differences exist aren't ones that concern me a lot. YMMV.
Well, LibreOffice is not "copyleft" licensed. It is distributed under a joint of licenses: GNU Lesser GPL v3+ and Mozilla Public License v2. The latter is not "copyleft".
As far as the Wikipedia definition of it goes, yes it is a "weak copyleft", but that basically means that the license on code and patents is liberal, while license on trademarks is not. That is to prevent bottom feeders to take for example Firefox and LibreOffice, rebuild them with adware and malware and redistribute new installers as "Super Firefox", "LibreOffice Improved" or other misleading trademarks.
Yeah, to be honest I could just use OpenOffice.org from the pre-Oracle days and it'd still do everything I need for the occasional time I care about office documents.
I didn't realise it came down to licensing. Thanks for clearing that up. Perhaps both are important.
So what you're saying is, LO vs AOO is real-world testing that shows a conclusive defeat of "permissive" at the hands of "copyleft", in terms of appeal to the class of users who know the difference.
Instead, it shows that should copyleft people desire, they can severely damage permissive projects by simply not doing what their own license requires. They kind of prove the point that you have to force people to give back, because given their own tendencies, they won't. LO could have easily been a bilateral partner, instead of a simple consumer.
Even so, LO benefited greatly by having AOO to consume, as I have mentioned elsewhere. You simply can't ignore that nor hand-wave away it's significance.
I can see how LO benefited from the original OO greatly, but AOO? Sure, LO has ported a bunch of stuff from there, but they wrote a great deal more from scratch (unsurprising, since that's where all the dev resources were).
And, aside from abstract good feelings, what benefit would having two forks of the same product, with no clear difference in philosophy (like you have with the BSDs, for example), serve? Sure, the licenses are different. But if LO guys cared about more permissive licenses, they would have just dual-licensed their own codebase.
I think the problem there is that you are, literally, one of the AOO developers who spent his time commenting on blogs and forum sites, flaming people who were pointing out the Emperor's new security policy - while not lifting a finger to actually stop distributing security holes to your own users. It is possible you don't have the moral high ground here. Dab hand at flaming people, though, well done.
I still need OO because in the documents I work with at least, LO will not let me select certain frames. I see an awful lot of LO good OO bad comments here and elsewhere, but my user experience doesn't really conform to that pattern.
What I mean, is that on occasions when editing .odt documents I create frames to organize for example some mono spaced text in a box. With OO if I move the mouse near the border of the box I am reliably able to click and select to grab onto the frame itself, for example to resize it. With LO this is hit and miss for me, for some reason. From my user perspective, LO is happy enough to show me the frame but refuses to let me manipulate the frame. Not always, and not for all frames, but often enough to be (very) annoying.
OpenOffice was tarnished by Oracle when they bought Sun and some of that bad will was transferred to Apache when they got it.
OpenOffice has had unpatched security vulnerabilities with exploits in the wild for 6+ months at a time.
Users today are downloading OpenOffice thinking it is the best free office suite out there when it's had no feature upgrades since May 2014. They'd often find out that it didn't properly support something they wanted to do (like handling MS Office formats) and think they had to go buy MS Office. Meanwhile LibreOffice is sitting there ready to be downloaded for free and handles MS Office formats much better than OO does.
Sure, there are some people who are finding schadenfreude in it, but many of us want to see the confusion, security issues, etc go away so things can move forward.
I want AOO to die because it has the brand recognition. Frankly, AOO is slower, less stable, less secure, less pretty, and overall a worse product than LibreOffice.
LibreOffice should have been the successor to OOo, and it offends me that my non-technical friends still install AOO when LibreOffice would improve their lives so much more.
When AOO began and actually had a release, I was surprised. I expected it to fail sooner. It was a mistake from the beginning.
Anyway, none of this means the project is actually going to be retired. It's just a discussion around something that might happen.
[1]: http://www.mail-archive.com/dev%40openoffice.apache.org/msg2...