Waiting for Apache Open Office
lwn.net
lwn.net
I think the LibreOffice project should use this opportunity to rebrand themselves and build even more distance from OpenOffice and MS Office. Pick a name that's easier to pronounce, spell, and remember - simple English words without two vowels in a row. (What was that? Did you say Leebray Office? Libruh Office? Lib Roffice? Does this have something to do with Mexicans and liberals? We're capitalists here!)
Everyone who cares about "libre" software will still know it's libre without putting it in the name. Communicate that this is the office suite that successful people use, not the one that poor people use because they're poor.
I also think that while LibreOffice is relatively new, it name has already begun to spread among normal users. I think it be very unwise to throw that good reputation away and start over. I also happen to be quite found of their current name, LibreOffice. I think it's classy and easy to remember.
LibreOffice and the document foundation can't use the OpenOffice trademark, but that doesn't mean we can't. We should write a lot of articles pointing to LibreOffice and calling it The OpenOffice, and try to make LibreOffice the #1 search result for OpenOffice. Let's genericize the hell out of OpenOffice.
Trademarks should ideally serve to protect customers, so they get what they think they are buying. But this is one of the cases where the exclusive right to the trademark hurts customers.
If the community were to write a ton of articles pointing to LibreOffice and calling it anything, that would help brand recognition for whatever they're calling it. MindStream, Keen Suite, Able Office, The Expressive Toolbox, the open office called LibreOffice, ... (random ideas)
I think LibreOffice can and probably will absorb OpenOffice's mind share over time, but I don't think either one stands a chance to become the standard suite that everyone expects to find everywhere. Not as long as dull license terms that almost nobody cares about remain the core of their identity.
http://luxate.blogspot.co.uk/2010/10/agreeing-on-childs-name...
tl;dr trademarks in lots of countries, URL availability in lots of countries. This stuff is hard. "Why don't they just make up a new name?" Well, that explains why.
It seems English is the only exception.
Do they think its a better product? Better community? Brand loyalty? Money? Did they loose a bet?
I'm not trolling (ok - maybe a little with the last question). I just want to know the story here.
I'm a contributor to LibreOffice, but don't speak on behalf of the TDF. But IMO, I don't think LibreOffice devs would be at all happy with a rename to OpenOffice - that ship has long sailed and the history of obstructionism, poor quality in terms of builds, bad faith comments from leaders at AOO and an utter inability to maintain their own software has sullied their brand to the point it's in many ways a net liability to take on the OpenOffice name now.
LibreOffice has made literally thousands of code improvements and major refactoring is still ongoing. If anything the rate has gone up exponentially, to the point now where we are beginning to actually compete with Microsoft Office. We are already better at a number if things than the Microsoft product - we even produce more compact XML than what is generated by the latest version of Office. In fact, in terms of product direction you have a far better chance of influencing product features in LO and getting issues resolved than you'll ever have with Microsoft. Microsoft Connect is less than useless in so many ways, it's not terribly open and I've seen bug reports that are acknowledged and sat on for years - with LO, you can pay a reasonable amount of money to a third party to get the feature implemented or the bug fixed. You could never do this with Microsoft.
I think that the way Oracle went about things should be a lesson for any company or organization: Oracle cannot be trusted, and if they give you a code dump, if there is a competing fork then seriously consider whether they should be consulted and supported before you accept the "gift" from Oracle.
What exactly is it you think Oracle did wrong, w/r/t the ASF? They donated code and trademarks, and as far as I can tell, they delivered exactly what they promised. Beyond that , everybody involved in AOO knew about LO from the beginning.. there was no confusion on that point.
If anything, the biggest lesson for AOO should be more about over-dependence on paid developers from one company. But in this case, that "one company" was IBM, not Oracle. AOO was moving along quite well until IBM decided to yank their support and all the paid staff they had working on the project. But then again, everybody involved knew that was a risk, from the beginning.
Yeah, Apache was stupid for accepting the dog. But it's also good to be pretty damn skeptical anytime Larry offers you anything.
Oh, and well done to the Apache Software Foundation for falsifying the meeting minutes when a journalist came calling to ask what was up. https://www.theregister.co.uk/2017/04/28/apache_openoffice_n... Note that's not the "independent" project, but the board of the nonprofit itself doing that.
BTW, AOO will probably finally get 4.1.4 out this month - they've got a release candidate for Mac and Linux, just need someone to successfully build revision 1803945 for Windows. http://www.mail-archive.com/dev@openoffice.apache.org/msg313... http://www.mail-archive.com/dev@openoffice.apache.org/msg314...
> As of 2017-Apr-12 we have more than 214,000,000 million [sic] downloads and it is still at a consistent rate with ~100,000 downloads in average per day". That is 100,000 people every day who are downloading the output of a project that clearly lacks the development capacity to get important bug fixes out to users, much less understand and improve the entirety of such a massive body of code.
100k downloads/day is a lot.
I just installed LO 5.4 on HiDPI Windows. The LO home page (localized) advertised that "LibreOffice 4 (sic!) just arrived". Unfortunately it went downhill from there: the HiDPI support is kind of not really present. Icons have negative margins (are halfway visible) and what's even worse the hit targets are not where the icon lies! So you select item N from a list and item N-1 gets selected.
I get it that OO is now evil due to Oracle but LO is just not usable. I know, I know free software but advertising LO to my family and friends would be the quickest way of getting them to not to ask me any IT related advice. Which may be a good side effect in itself.
Edit: I just tried AOO. It works. Looks a little bit dated but the controls are where they should be and everything I tested works correctly.
Just to be clear-- are you saying that OpenOffice correctly handles HiDPI in its GUI while LibreOffice does not?
When you say "the controls are where they should be," do you mean that the icons in OpenOffice are both properly sized in the GUI window as well as having the correct hitboxes?
I'd love to see some screenshots of the two side-by-side if you have some.
It's probably upscaled but it works correctly.
> When you say "the controls are where they should be," do you mean that the icons in OpenOffice are both properly sized in the GUI window as well as having the correct hitboxes?
Yes.
> I'd love to see some screenshots of the two side-by-side if you have some.
Please check HiDPI bug list for screenshots. After a little bit of tweaking it turns out that disabling "Full OpenGL rendering" in settings fix these issues. Now LibreOffice looks and works just like AOO.
The problem you are having with the icons and clicks mis-registering is due to OpenGL rendering. I had the same problem. Go to Tools-Options-View and disable OpenGL rendering, should fix the problem.
Looks like there is some problem with the Intel drivers and LibreOffice, not sure.
* Dell Precision 7720 Nvidia Quadro P5000, Driver 385.08 * Windows 10 Pro, 64-Bit, Version: 1703, 15063.502
You are awesome for this solution!
It was also quite telling that when a security vulnerability in 4.1.1 was discovered, at least when checking the wayback machine the earliest security announcement was posted at April 29th 2015 ( http://web.archive.org/web/20150429084734/https://www.openof... ) and while multiple snapshots have been captured since ( http://web.archive.org/web/20150501000000*/https://www.openo... ) it took until November 1st 2015 for the recommendation to change from "delete the .dll" to "upgrade to 4.1.2" ( http://web.archive.org/web/20151101015732/https://www.openof... ). At least until September 24th 2015, version 4.1.1 was still the latest version and the recommendation to delete the .dll was still present ( http://web.archive.org/web/20150924061922/https://www.openof... )
The OpenOffice brand is still very well know, and unfortunately the current situation might trick average joes to download and outdated unsupported build, while the very active LibreOffice is around the corner.
I don't remember if I ever convinced them, but it became even clearer to me that OO's development is the dried husk of its former self.
Even though I'm one of the minority that understands the nature of the fork, and the current status of both projects, as some LO developers have noted here it would not be palatable to some of them to try to flip back to 'OpenOffice' as a project name. I don't believe educating users of OpenOffice about the name change is an intractable problem, but, again as some LO contributors have noted here, it's not necessarily LO's role to do this, nor is it something they necessarily could do. But it is something the AOO team could do, and, as per TFA, it's probably now the right thing to do.
> How would the Lo people feel about a simple redirect from Oo.o-> Lo.org w/some kind of banner welcoming users, and some kind of upgrade path from Oo->Lo?
That would be a better option, IMO. LibreOffice not being well known would be helped if such a move were done. Changing the name of LibreOffice to something else would be a longer marketing exercise, and if LibreOffice does want such a change, I'd rather have it become something entirely new with "Office" as a prefix or suffix than using the name OpenOffice (I personally don't like it because of the history associated with it).
Isn't that why it's "Apache OpenOffice" because they can't call it just "OpenOffice", similarly Siemens have a mark "HiPath OpenOffice".
Eg https://trademarks.ipo.gov.uk/ipo-tmcase/page/Results/1/UK00... a dead application for "openoffice" from Oracle, expired in 1997 on non-payment of renewal.
[A modicum of research ensues ...]
Seems, https://www.tmdn.org/tmview/bookmark?s=ju0uijpgg5uupublble76..., there's a valid mark in Canada, Switzerland, and Russia owned variously by Sun, Oracle, Apache.
It's very generic IMO; a poor mark even if it could be secured.
The value of open source is that the code is there for anybody that wants to contribute, or fork in case people find its maintenance unsuitable. And even if the project dies, the code remains there for anybody to pick it up later, in case it has value. It has happened before, it can happen again. On the other hand most open source volunteers are unpaid and have no contractual obligation.
So tell me who is the criminal?
The Apache Foundation? But that's silly, since this project might have been simply dumped on GitHub. The few volunteers that tried pushing some fixes since last year? Oracle for donating the project? Speak for yourself in this case, because I'd rather have companies release their stuff, instead of locking it away.
This is an upsetting entitlement complex show here. If you want to contribute, then do so, otherwise GTFO.
It's not legal for me to set up a shop selling known faulty electronics, or even give them away for free unless I make clear I know they are faulty. I wonder if this should be extended to tech.
I'm not saying they have to take the code down entirely, but they could pop a big warning on the download page at least. They are knowingly having thousands of downloads a day of software they have known is dangerously broken for over a year.
All non-trivial software has security bugs. To reiterate, every program you've ever heard of has had and will have serious security bugs. If no bugs have been published, then either nobody is looking or someone is keeping them secret.
I agree that Apache has behaved irresponsibly.
However contributors _and Apache Board Members_ chimed in to announce that Apache OpenOffice was healthy and would continue. They managed, weeks laters, to spit out a release containing almost literally just one bug fix and then nothing since. So long as Board Members pretend everything is fine the Apache process won't trigger. This has become about the ego of Board Members versus the interests of the wider community.
Warranties are not the same as being misleading.
------
> > One thing has definitely changed: the last time LWN wrote a critical piece about the project, we had some AOOers drop by to repeatedly complain about what a bunch of meanies we all were for criticizing them instead of volunteering to help them out. This time? Crickets...
> ...or they are not LWN subscribers, and nobody posted a SubscriberLink somewhere they could find yet.
> Last year I had posted a SubscriberLink to a LWN article about AOO to HN; I'm doing it again right now, so if they read HN and my post reaches the front page there, we might be able to see them here again.
-------
I'm not sure if HN is really supposed to be used like that..
Or maybe we're tired of wasting our time on this public drama, none of which - at the end of the day - helps get any software shipped.
I'm just sad that I don't personally have more cycles free to contribute to AOO. I have a bunch of stuff I want to work on, but my life is so chaotic right now that it doesn't happen. So I guess I share in the blame for the lack of AOO progress. I wish I had something more to tell you guys, but it's a project that's driven pretty much 100% by volunteers these days, and those volunteers do what they can while maintaining their day jobs, etc.
http://mail-archives.apache.org/mod_mbox/openoffice-dev/2017...
So yes, development on AOO is absolutely continuing. Is the pace slower than we'd all like? Absolutely. Would we welcome newcomers to help out? Absolutely. In the meantime, the volunteers that are around are doing what they can.
I really wish people would let this whole supposed "AOO vs LO rivalry" thing go. At this point, I'm pretty sure the bulk of the AOO contributors don't see LO as a rival, and mostly don't care what LO does, so long as people affiliated with LO aren't running us down in public. The two projects will probably never re-merge, so it is what it is. They do their thing, AOO does its thing. Such is the way of things.
Given that most of the energy is with one project and not the other, and codebases have diverged so much that one can't accept patches from the other, why is it beneficial to have both? Who does OOo serve that LO doesn't? And why would a volunteer choose to work on OOo and not LO if they weren't on the OOo side of the fork to begin with?
It sounds like you expect some sort of objective, definitive answer to those questions. I don't think there are any such answers. A lot of people use AOO, a bit fewer contribute to development. I can't explain the motivations of any of those people, outside of myself. I can't even explain why so many of the "old school" OO developers "sided" - so to speak - with LO, other than just the timing of the thing.
But I came into AOO after a lot of the weird stuff had already happened, that created all this bitterness and resentment between the two projects. I'm a bit of an outsider as it were. So I have a hard time understanding why there even is all this acrimony and tension at all.
So I don't have a lot of good answers for you. All I can say is that, for myself, I gravitated to AOO for a few reasons:
1. I prefer the Apache License
2. I was familiar with, and liked, the ASF way of doing things.
3. I started out with high hopes to leverage what seemed like a natural synergy between AOO and other ASF projects. In particular, some of the Semantic Web tech (Jena, Fuseki), and all the "big data" projects. It just felt natural to me to want to work on AOO since there was a certain amount of overlap in membership between these various projects, etc.
Sadly, I got so bogged down with other activities, that I have never found time to really contribute to AOO and work on a lot of the things I had hoped to work on. I still cling to some of those ambitions, but other priorities keep getting in the way.
Sorry for the long, rambling semi-answer. I'm tired and just kind of going "stream of consciousness" here.
edit: See the second and third subsections under this: https://en.wikipedia.org/wiki/LibreOffice#History. LibreOffice was the fork motivated by dissatisfaction with Oracle. OpenOffice code was donated, along with trademarks, etc. to the ASF.
When OpenOffice pushed back (and after they completely shafted Kohei's effort in contributing a solver for Calc) they in essence told the wider development community that their contributions weren't welcome and forced the creation of The Document Foundation to govern the development of LibreOffice. The TDF then worked out how to be welcoming, lowered barriers to entry for new contributors massively, put in infrastructure like gerrit and opengrok, majorly simplified the build process and generally showed a lot of love to their developers. At this point it was game over for Apache OpenOffice, and now everyone contributes their code to LibreOffice. Which is an utter no-brainer really - why send your code to a bunch of people who arrogantly think your efforts are by default crap as you have no experience with the codebase?
It's frankly all kind of old news for most LO devs. We're far more interested in improving LO and competing against our real rival, which is Microsoft. Every now and then someone might make a small comment about AOO, but we don't encourage a lot of criticism of AOO because we have enough to do than spend any effort on ill-will towards Apache. We want to drive positive change and win by being awesome, not "win" through negativity :-)