Is OpenOffice dying?

[obswob]

I started using it for a fairly big document and every single day I'm running up against bugs and crippling design limitations. I'm getting used to working around them. But I notice some of these bugs have been reported 10 years ago and still aren't fixed. I know there's a new major version due in a month, but that doesn't explain 10 year old bugs. Perhaps there's more development effort on LibreOffice than OpenOffice? Is that the direction of the future?

[Hagar Delest]

Not at all.

Bugs are tackled depending on the manpower and also on the good will of the developers (or the fun they find in working on them). Same for LibreOffice. Just try it and if it has less bugs for you, then go for it.

Most devs are more interested in implementing new features (their code is hence more "visible" to the community). Fixing bugs has less impact (personal marketing point of view), even if that's what most users are waiting for.

There have been some discussion on the dev mailing list about bugs already, you can browse the thread (link on the AOO support page). But don't bother asking, you'll get the standard reply: "you want a fix? then just code it". That's a drawback of open source community I guess: you can't really influence the way it works (even if you count votes in the bugzilla), it all depends on what the devs want to do for their own agenda.

[jrkrideau]

I started using it for a fairly big document and every single day I'm running up against bugs and crippling design limitations. I'm getting used to working around them. But I notice some of these bugs have been reported 10 years ago and still aren't fixed. I know there's a new major version due in a month, but that doesn't explain 10 year old bugs. Perhaps there's more development effort on LibreOffice than OpenOffice? Is that the direction of the future?

Just how big and how complicated is the document? OOo is designed to handle fairly large and rather complicated documents but I don't think any plain word processor can do everything.

What are some of the bug and crippling design limitations. Some users may have work arounds or in some cases you may just be out of luck. It's hard to tell without some specifics.

[orangeli]

I know there's a new major version due in a month, but that doesn't explain 10 year old bugs. Perhaps there's more development effort on LibreOffice than OpenOffice? Is that the direction of the future?

Don't count on version 4.0 to be less buggy - it is a leap from 3.4.1 without 3.5, 3.6, 3.7, 3.8, 3.9 in between. It will also have a major UI change - the so called "sidebar".
From my personal experience the OpenOffice versions from the Sun era are the least buggy.
Also from my personal experience LibreOffice is highly not recommended.

[obswob]

Don't count on version 4.0 to be less buggy - it is a leap from 3.4.1 without

Maybe the developers will be more enthusiastic about 4.0 because it's new and don't want to waste effort on the old one. I know from my own experience that it's a real drag fixing bugs in an old version when the whole feature has already been rewritten in a newer one.

jrkrideau: You shouldn't have asked! 100-200 pages. Problems like repagination not updating even when exporting to pdf, being unable to change the style of a section number independently of its text before the number - thus having 3 different fonts in a heading, having to keep deleting the user profile because it regularly causes a whole slew of bugs like being unable to save or copy and paste, references not being formatted correctly then randomly fixing themselves days and many edits later, footnotes being too rigid in their formatting, pictures extending off the bottom margin of a page (this is a biggie. I don't want to manually insert 100 page breaks and maintain them on every new version). Jumping to the end of the document when inserting a picture, autosave taking maybe 30s and the program freezing up while it's happening, Windows Explorer windows freezing up until you close the OO Quickstarter.

[orangeli]

Enthusiasm is irrelevant.
Renovating the penthouse won't do if the building foundations are shabby.

[Hagar Delest]

Enthusiasm is irrelevant.
Renovating the penthouse won't do if the building foundations are shabby.

+1.

Developers see AOO as a way to have fun by contributing code that will increase their status (that's just human). But users of a productivity suite are more interested in stability and robustness than new features, this is the general pattern that clearly appears from the questions in forums.

About the inclination of the developers, see this discussion on the dev mailing list: A question about existing practices (rather long but very representative of the spirit of the mailing list for other similar topics).

[acknak]

I suspect that unless the project finds a way to pay some core developers it will become stagnant. OOo is just not the kind of project that attracts volunteer developers.

First, I don't think people with coding skills care much about productivity software--at least they aren't interested in the kind of grunt work that makes productivity software really good. Second, coders aren't interested in the user base (non-tech users). Third, the code and structure of OOo is a swamp. Parts of it are great, but much of it is old (like Babylon: layer upon layer of old), complex, non-standard, obscure, and difficult to work with. It takes a huge effort just to get to the point where you can start to hack at it. All this isn't going to attract volunteer developers.

I give LibO a lot of credit for trying to address some of these: removing some cruft, trying to move to some newer coding tools, etc. but LibO seems to have it's own problems.

What I've seen of the sidebar looks like a marginal improvement at best. Maybe, once the push to 4.0 is finished, some of the core problems can be addressed.

[orangeli]

What I've seen of the sidebar looks like a marginal improvement at best.

That's an understatement.
Professional office applications shouldn't have the user interface of an iPhone application.

[RGB]

IMO, generalizations are always a bad thing: I think it's wrong to say from outside what "developers" do or do not or what they like or don't.

For example, possibly for 4.1 there will be a HUGE addition to AOO to easy accessibility to the program for disabled persons. An enormous work, running from several years that redesign almost from scratch the whole graphic layer of AOO addressing several issues is being completed right now and will be included on future releases. There are tons of improvements on document import fidelity being cooked right now... and that's not mentioning the certainly not glamorous changes "under the hood" like changes on the APIs, directory structure for installed files, extension management, etcetera.

A note about the "marginality" of sidebar: while I don't think I'll use it on Writer beside the stylist and navigator decks, the properties deck is an enormous improvement on Draw and Impress, allowing the user to tweak objects on real time without the need to open again and again menus and dialogue boxes. Also, remember that in principle it's possible to easily add decks through extensions so the sidebar is far more of what you'll see on 4.0.

[acknak]

IMO, generalizations are always a bad thing: I think it's wrong to say from outside what "developers" do or do not or what they like or don't.

Sure. I know what you mean, and I've never had a conversation with any OOo developer at all, so I could be completely wrong. However, I have tried to get involved with some coding in OOo and those are some roadblocks that I hit so maybe there's at least some reason for considering those things.

I should also say that I have tremendous respect and appreciation for everyone who contributes to OOo. I just wish that there was some more focused management of what gets worked on. Rob's garden analogy (see Hagar's link) is helpful: if you have more garden than you can manage, it becomes even more critical to work with a plan.

A note about the "marginality" of sidebar: ... the properties deck is an enormous improvement on Draw and Impress, allowing the user to tweak objects on real time without the need to open again and again menus and dialogue boxes. ...

Oooh, I missed that part. I was only playing with it for Writer. That would be a big improvement indeed. I'll have to give it a try under Draw.

Thanks for pointing that out!

[jrkrideau]

jrkrideau: You shouldn't have asked! 100-200 pages. Problems like repagination not updating even when exporting to pdf, being unable to change the style of a section number independently of its text before the number - thus having 3 different fonts in a heading, having to keep deleting the user profile because it regularly causes a whole slew of bugs like being unable to save or copy and paste, references not being formatted correctly then randomly fixing themselves days and many edits later, footnotes being too rigid in their formatting, pictures extending off the bottom margin of a page (this is a biggie. I don't want to manually insert 100 page breaks and maintain them on every new version). Jumping to the end of the document when inserting a picture, autosave taking maybe 30s and the program freezing up while it's happening, Windows Explorer windows freezing up until you close the OO Quickstarter.

You should not be getting anything like all those problems in a normal odt document. You may have some nasty system's level problems or perhaps a slighty corrupt AOO installation...

The profile problem is a long-standing one, and the little I know about it suggests that the problem about curing it is that it cannot be reproduced reliably so troubleshooitng it is almost imposssibe. That said, I've been using OOo & AOO for something like 7 or 8 years on a variety of machines--Windows XP, & and Ubuntu and have only had to do a reset once under Windows 7.

I had not heard of any problem with Quickstarter and Windows Explorer but, also I have not had Quickstarter installed in years .

I don't usually work with large documents, 40 pages is large for me but others here write books with it, so I don't think your problems are really typical.


I'll send you a PM later.

[RoryOF]

My experience with two plain text files (book format), each of 300+ pages being built over five years, each with about 3500 edits, is that they hold their formats well, no troubles with their footnotes, endnotes or comments or with the User Profile. No pictures or tables - they are narrative text. I have no experience of running on Windows 8 - I use Windows 2000 or XP, when I must use Windows and couldn't get Win 8 off my new laptop quickly enough (now running Xubuntu on it).

[crusader]

Bugs are tackled depending on the manpower and also on the good will of the developers (or the fun they find in working on them). Same for LibreOffice. Just try it and if it has less bugs for you, then go for it.

+1
I have found LibreOffice to be an exceptional product...

[RGB]

The profile problem is a long-standing one, and the little I know about it suggests that the problem about curing it is that it cannot be reproduced reliably so troubleshooitng it is almost imposssibe.

But just "almost" In fact, most if not all the problems that produce corrupted profiles will be solved on 4.0

That said, I've been using OOo & AOO for something like 7 or 8 years on a variety of machines--Windows XP, & and Ubuntu and have only had to do a reset once under Windows 7.

I'm using StarOffice/OOo/AOO since 1999 and never needed to reset the profile on a final release.

I don't usually work with large documents, 40 pages is large for me but others here write books with it, so I don't think your problems are really typical.

I have some 400+ page pure text documents, one 300+ page with ~70 pictures and dozens of tables and no problems.

[RoryOF]

RGB's experience, that of jrkrideau, acknak and my own experience as reported above would lead me to suggest that stability of the host system is the first essential for a reliable Open/LibreOffice installation.

[acknak]

... most if not all the problems that produce corrupted profiles will be solved on 4.0 ...

Wow. That will be a huge improvement.

Can you elaborate a bit on what is changed to have such an effect?

[RGB]

... most if not all the problems that produce corrupted profiles will be solved on 4.0 ...

Wow. That will be a huge improvement.

Can you elaborate a bit on what is changed to have such an effect?

Not really: I do not understand the technical part of the problem. It was mentioned on the dev mailing list that there is a fix for the problems that appears on upgrades, for example the random crashes that several users had going from 3.3 to 3.4.x will not be present on a 4.0 update. More info here

Bug 121625 - Tracking issue for AOO 3.4 startup related crashes

Marked as "resolved-fixed"

[acknak]

Ahh, ok. Got it. Thanks.

[JoseFrancisco]

There may be glitches in terms of the software updates. However, for us in our setting, Openoffice is a good option, if you do not want to use Microsoft tools.

[CaptainPuget]

This question (is OpenOffice dieing?) came to mind as i started reading the documentation (http://wiki.openoffice.org/wiki/Documentation/DevGuide). I just downloaded OO for the first time, and thought I'd make an exception to my general rule, and look at the documentation. A fair amount seems quite out of date - dialogs with "Star Office" in the title, for example, as well as rearranged buttons and modified text. The last time some of the pages were touched was 4 years ago. I don't have an issue with the documentation per se, but I'm wondering if the volunteers are getting a little tired of spending time on the project.

Maybe I'd be better off by just not looking at the doc!

[RGB]

This question (is OpenOffice dieing?) came to mind as i started reading the documentation (http://wiki.openoffice.org/wiki/Documentation/DevGuide). I just downloaded OO for the first time, and thought I'd make an exception to my general rule, and look at the documentation. A fair amount seems quite out of date - dialogs with "Star Office" in the title, for example, as well as rearranged buttons and modified text. The last time some of the pages were touched was 4 years ago. I don't have an issue with the documentation per se, but I'm wondering if the volunteers are getting a little tired of spending time on the project.

Maybe I'd be better off by just not looking at the doc!

Try here: INTRODUCTION TO DEVELOPMENT

[CaptainPuget]

So is all the openoffice.org material pre-apache (but maybe post-Oracle), and I should just look at openoffice.apache.org? Is this related to the observation that some excellent suggestions in Talk,
made 5 years ago, seem not to be reflected in the Page? Oy.

[CaptainPuget]

Is there a more up-to-date version of http://wiki.openoffice.org/wiki/Documen ... ASIC_Guide somewhere?

[Villeroy]

Is there a more up-to-date version of http://wiki.openoffice.org/wiki/Documen ... ASIC_Guide somewhere?

Nothing changed. Basic remains the same and the API is documented on the internet.

[Harshada]

Hi,

Oracle has just recently given up the development on OpenOffice, they are not involved in any further developments. Before that, there were some internal disputes, which have caused some of the developers to leave the team and start a new derivate of OpenOffice, which we nowadays know as LibreOffice. OpenOffice is pretty much dying right now, since its team is really broken. If you want something future-proof, you should go with LibreOffice.

[RoryOF]

Oracle has just recently given up the development on OpenOffice, they are not involved in any further developments. Before that, there were some internal disputes, which have caused some of the developers to leave the team and start a new derivate of OpenOffice, which we nowadays know as LibreOffice. OpenOffice is pretty much dying right now, since its team is really broken. If you want something future-proof, you should go with LibreOffice.

This misinformation is two years out of date. The code transfer to Apache took place in mid 2011, since when Apache have released v3.4 and 3.4.1, of which there have been over 50,000,000 downloads; v4.0 is due for release - projected date end of June 2013. It is supported by an active community of Developers.

[orangeli]

This misinformation is two years out of date.

To avoid such misinformation there should have been release of 3.5-3.9 with bug fixes every 2-3 months (and even more frequently if the incremental update mechanism was implemented and there was no need to download the entire executable for install).
This is the best way to show OpenOffice is alive.

[robweir]

This misinformation is two years out of date.

To avoid such misinformation there should have been release of 3.5-3.9 with bug fixes every 2-3 months (and even more frequently if the incremental update mechanism was implemented and there was no need to download the entire executable for install).
This is the best way to show OpenOffice is alive.

There are different opinions on that. Firefox, for example, had a "rapid release" schedule that annoyed users and was partially responsible for driving them to Chrome:

http://www.forbes.com/sites/adriankings ... x-updates/

http://hftechgroup.com/index.php/7-news ... g-too-fast

For corporations especially it is better to have larger upgrades at larger intervals than frequent micro-updates. Look at Microsoft, for example. They have a major update every 3 years and a service pack in the in-between years.

The release pace for Apache OpenOffice -- twice a year -- is in the middle. In the end our users want to spend time doing their task, their work, their school work, whatever. Maybe some obsess over their tools, but for most users OpenOffice is just a tool, a means to an end, and having a stable, familiar tool that gets the job done is golden. There is little pleasure in getting a new knife and fork every 2 months, with new bugs, except for the small minority of technologists who find updates fun rather than risky.

[orangeli]

I'm sure the lack of frequent updates drove users to LibreOffice.
Of course extremes are bad - silence for half a year / update every week.
The above arguments are more of a reason to eliminate the need to download a three-digit MB file for update rather than for infrequent updates.
Micro$oft was (past tense) a good example.