LWN.net Logo

Few really new features in AOO (incubating) means nothing

Few really new features in AOO (incubating) means nothing

Posted Apr 26, 2012 14:01 UTC (Thu) by cesarb (subscriber, #6266)
Parent article: Meeks: A LibreOffice/Apache OpenOffice Comparison

Having few new features (or even no new feature at all) in 3.4 would be expected, and in fact it is a good thing. They need a "baseline" release to base their work on, and need it fast (the longer they wait, the greater the chance of a critical security bug in 3.3 they cannot distribute a fix to). So their focus would be in releasing 3.4 as soon as possible, and not in adding new features to it (or even fixing bugs and cleaning up the code).

A real comparison will be possible only with the next major release following AOO 3.4 (be it 3.5 or 4.0).


(Log in to post comments)

Few really new features in AOO (incubating) means nothing

Posted Apr 26, 2012 21:59 UTC (Thu) by josh (subscriber, #17465) [Link]

That argument would carry more weight if AOO didn't bring that pain on themselves via the Apache licensing policy of refusing to use GPLed/LGPLed components.

Few really new features in AOO (incubating) means nothing

Posted Apr 26, 2012 23:00 UTC (Thu) by cesarb (subscriber, #6266) [Link]

> That argument would carry more weight if AOO didn't bring that pain on themselves via the Apache licensing policy of refusing to use GPLed/LGPLed components.

That policy is no fault of AOO; as far as I have heard, it has been Apache policy for a long time.

If we were trying to place blame, the fault would lie with the ones who chose to drop the giant ball of code in Apache's lap, knowing fully that it would mean creating a need to purge parts of that code.

The Apache OpenOffice (incubating) community had no choice in this matter, since it did not exist yet (it obviously started existing only after the code was thrown to Apache). They have to do their best with the preexisting situation: the strict Apache policy, the inherited bundle of code and artifacts with their licenses and grants, and the resulting intersection.

IMHO, their best code of action (if they decide to go on separately instead of merging with LO) would be to release a clean working 3.4 as a base as soon as possible, and go on from there. From what I have seen, this is the course of action they are following.

Few really new features in AOO (incubating) means nothing

Posted Apr 28, 2012 15:25 UTC (Sat) by drag (subscriber, #31333) [Link]

> That policy is no fault of AOO; as far as I have heard, it has been Apache policy for a long time.

Yes: This is a not a unusual issue that crops up with all Apache projects. They are very strict on the licenses they are willing to accept or re-license to.

> If we were trying to place blame, the fault would lie with the ones who chose to drop the giant ball of code in Apache's lap, knowing fully that it would mean creating a need to purge parts of that code.

This seems like a inappropriate project for Apache to take on, then. Just because the ball was thrown at them doesn't mean that they are obligated to run with it. Maybe the devs on the Apache project would better serve themselves by placing their contributions under the auspices of the Libre project.

That's not strict, that's rigid

Posted Apr 28, 2012 20:15 UTC (Sat) by coriordan (guest, #7544) [Link]

> They are very strict

Sounds more rigid than strict.

If a policy blocks the goal, the policy should be discussed. (That's assuming that the goal of AOO is to improve the free software options available to users of office suites.)

Take RMS for example. No one doubts that software freedom is his goal. He has policies about using GNU licences and GNU contributors assigning copyright, but in limited circumstances he recommends non-GNU licences and accepts code into GNU packages without copyright assignments. He puts the goal ahead of rigidly following policy.

Before AOO, there was little worry about the motivations within the Apache project. But there are also no illusions about the motivations of Oracle Corp and IBM. Each has a lot of influence in AOO. If a decision is bad for software freedom but good for Oracle or IBM, and the decision is supported by rigidly following a policy, then "we're really strict" can be a convenient fig leaf.

OO development processes mean everything

Posted May 3, 2012 23:44 UTC (Thu) by brianomahoney (subscriber, #6206) [Link]

I didn't realise until I read your comment for the third time is that the way you think and react is exactly what is wrong with the OO, but not the LO culture ... your culture is one in which code bases need to be PROPERLY maintained and needed new features come from the Marketing Department, which the get by talking to self selected senior users.

OO has been incubating for decades, always trying to catch up with MS Office.

Bound to fail!

LO has what any good software needs, a vibrant developer community that want to do GOOD, NEAT things and ARE, without waiting for some droid's permission.

OO development processes mean everything

Posted May 4, 2012 12:41 UTC (Fri) by cesarb (subscriber, #6266) [Link]

> I didn't realise until I read your comment for the third time is that the way you think and react is exactly what is wrong with the OO, but not the LO culture ... your culture is one in which code bases need to be PROPERLY maintained and needed new features come from the Marketing Department, which the get by talking to self selected senior users.

What.

I completely fail to understand how you could get *that* from my comment.

Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds