Apache OpenOffice Community Forum

Hagar de l'Est wrote:

pz1 wrote:I sincerely hope they will concentrate on bug fixes

I fear it won't be the case. The first move being to include all the Novell features in th main code (contrary to what was said in their site BTW), I'm rather worried about the direction they are heading.

When you speak of the Novell features, do you mean the stuff that's in Go-oo? I've tried their releases and didn't notice any stability problems... I think some of the changes they made are bug fixes, in the sense of eliminating performance bottlenecks.

Say what you like, if OOo is going to be used by more and more people, the interface with MS Office will have to improve, for the simple reason that so many people who use it will do business with people/companies using MSO. And as go-oo seems to be better with that, what's the problem with using it?

There are some glitches in the go-oo version (especially with Calc IIRC) ant more on GNU/Linux (since it's the standard version shipped with most of the distros). There quite some topics about such problems being fixed by the installation of the vanilla version (here and in the Ubuntu forums).

As for the compatibility with MS Office, I consider it's too dangerous to work with something else than MS Office when working with their file format. There have been some critical issues with tables in .doc for example, leading to data loss. Who can say that it won't happen again (some cases have not been fixed BTW!). The point by using OOo is to use ODF (and styles). Like everyone uses the same length for the meter (even if some like to use the imperial system, but it has made some space ships crash IIRC).

Bad things I've seen on go-oo include: the utterly broken kde4 integration, the "enhancements" to pdf export that makes exported links unusable, the great "use hardware acceleration" options that makes Impress to crawl when not to crash... They also had a tendency to deliver those patches without too much QA: for example, when they introduced the kde4 "integration" on a "stable" (for them) build, Novell's go-oo gave to kde4 users empty docs (already solved), crashed with certain extensions (still unresolved, only workaround is to uninstall kde4 integration...) and a loooong etc. Perhaps the most funny of these "enhancements" with low QA was when they decided to change the menu to select number of rows and columns for tables on Writer: it was not possible to reach that menu if used from the Tools toolbar. This last one was quickly solved and was not really importan, the point is that they delivered that obviously broken "enhancement" as part of a "stable" build.
So I share Hagar's fears, specially after "exchanging" a couple of times these (and others) "issues" with Mr Meekes on different blogs (I mean, the few times he was willing to share his comments...).
An Office suite is not a toy, it is not something "fun to hack on" (Mr. Meeks words): an office suite is a tool that must be stable, something that go-oo never was.

Hagar de l'Est wrote:The point by using OOo is to use ODF (and styles). Like everyone uses the same length for the meter (even if some like to use the imperial system, but it has made some space ships crash IIRC).

Yep , Rob Weir's opinion about ODF is also my opinion :

But if forced to take sides, then my loyalties are clearly going to fall to to ODF rather than to any one implementation. The ODF open standard transcends implementations and code bases. It is bigger than any one product. ODF is what enables the user to have choice.

I also remember the Katarina/Fema and Thailand/tsunami disaster :

Still not convinced? Consider this: The Bangkok Post reported that file format incompatibility allegedly hindered government recovery efforts after the 2004 tsunami. Since then, the Thai government has made open file formats an immediate priority. But open standards have not been relegated to “developing” nations. In the U.S., for example, victims of Hurricane Katrina who requested aid on the FEMA Web site had to use Internet Explorer. Got Linux? Sorry, out of luck.

Well I suppose it is because I do not push Calc that hard (simple P&L stuff for delivery to my CPA), and perhaps because I use exclusively Gnome for the production desktop, but I have used Go-oo for my production work for the last couple of years under Linux and my experience has been nothing but positive.

I'm aware of the problems with the go-oo version, but it's possible to make a hybrid version, wouldn't it, where the code is compared between the two versions and where the best parts are combined, though that will be very hard to achieve.

Of course, it would be the best! But how the QA will be managed if Oracle keep the Issue Tracker? Use another system like the Ubuntu launchpad? At best the IT database may be copied (depending on the license of its content), else, a new list of bugs? Who will be the testers? How hard the policy will be about showstopper bugs (see the examples given above, the PDF one being a rather critical one I've seen several times in forums)?...

Lose the issue tracker. It's a pain.

The issue tracking system for LibreOffice will be maintained on the Freedesktop.org platform - Bugzilla again.

For go-oo the main system has been the Novell bug tracking system, just like with OO.o all you need do is create an account and you can open issues.

Ubutnu maintains their tracking system under Launchpad, just like OO.o all you need is an account and you can open issues.
The ubuntu system has the ability to link issues from external systems, so it could track an issue that needed to go up stream to Go-oo or to OO.o.

The OO.o qa group has never been interested in tracking issues outside of the OO.o system - not their problem has been their mantra. I suppose they have a case for that.

I can tell you that from the numbers I've looked at the majority of issues that are opened on these other systems end up being upstream OO.o issues.

DrewJensen wrote:The issue tracking system for LibreOffice will be maintained on the Freedesktop.org platform - Bugzilla again.

Well great for end users like me who are already totally confused in this OO-jungle. I am certainly not waiting for yet another useless bug reporting system. Have a serious look at the Issue in my signature. The problem was clearly reported a couple of years ago. Only 2 people including myself cared to vote on the issue so it would get some attention. So I am rather pessimistic on the quality improvements in the future.

Can someone speak to the issue of extensions? Will OOo extensions be usable in LO without modification by the developers or will developers have to write two versions of the same extension one for OOo and one for LO?

For example: I depend on Zotero an awful lot. I simply can't move to LO if it doesn't play nice with Zotero.

Well, I saw a post from someone that has started testing extensions and so far everything they tried, sorry don't have specific names, work without problem.

Remember that on day one the code is really the same - or awfully close to it, so extensions should not be a problem - going down the road, well...my crystal ball gets a bit cloudy...(no pun intended)

DrewJensen wrote:Well, I saw a post from someone that has started testing extensions and so far everything they tried, sorry don't have specific names, work without problem.

Where is this post? Is there an LO-specific forum? (I've looked on the foundation page without success.)

Thursday, September 30, 2010

Extensions that are broken by the change to 3.3 code, will be broken in LO as they are in OOo. I have both LO (3.3m6) and OOo (3.3m9) on iMac.

Phil

It was in the mailing list discuss -at- documentfoundation -dot- org

I'm using Oo for all my work, recently for writing a book. And I'm making a Linux distro, for what I provide own Oo .tgz packages.

Generally, I think it was necessary to continue with the office program by an independent project. And soon I'll make packages for LO.

One of the reasons why SUN lost many customers, was that they 'thieved' the contributions and work of people. People aren't so stupid as they think not to see these wicked practics. Oracle seems to be worser even (but I guess this at the end will bring them downwards too). Probably they would try to privatize it. If they really were interested on keeping its open nature, then, now as the decision was made, it's supposed they contribute on the foundation and also give free the name Oo; the contrary would confirm that this wasn't them intention (and a-posteriory would justify the move even more). Very likely, the comercialization of Oo would not have many sucess, as it hadn't StarOffice - but if they think different, they should bring forwards StarOffice, for what they want 2 office programs ?

Under SUN, the development of Oo gone very slow. Under Oracle, I dont know about any progress yet.

I suppose that all relevant developers move with the office.

However, I agree with that the foundation will have to make efforts to bring forwards the program. Otherwhise it's probable that another firma (perhaps IBM) makes a fork and do this better.

Just to let people know the DocumentFouncation mailing lists are now being archived on the Nabbl web server -

http://documentfoundation.969070.n3.nabble.com/

Currently this only includes 3 mailing lists, Announce, Discuss and Mirrors.

Using Nabble you can read the archives and you can also subscribe to a mailing list via Nabble which allows you to post directly to the Mailing list without actually having to catch all the email in your inbox, rather you can use the web interface in a fashion similar to a web forum.

TerryE wrote:Perhaps Clayton can clarify when he himself knows.

Just a quick reply here - I'm reading the threads, but... I cannot comment on the topic. Oracle has issued a statement and you can find that statement quoted in many places now. There's nothing more I can say about it unfortunately, and I have to remain on the sidelines.

About the name LibreOffice...
→ Agreeing on the child's name ... a simple task?
by Christoph Noack, 2010-10-03

<http://luxate.blogspot.com/2010/10/agre ... -task.html>
<http://planet.documentfoundation.org/>

ccornell wrote:Oracle has issued a statement and you can find that statement quoted in many places now.

C, that might be true, but I can't find it on the Oracle or OpenOffice.org sites. It seems strange to issue a statement that isn't easily accessible. Have you a URL for a reference source of this?

I have no direct link. The quoted text that you see in the media is Oracle's statement.

eg: http://www.crn.com/news/applications-os ... *.ecappj01

Oracle as reported by CRM wrote:Oracle is investing substantial resources in OpenOffice.org. With more than one hundred million users, we believe OpenOffice.org is the most advanced, most feature rich open source implementation and will strongly encourage the Open Office community to continue to contribute through http://www.openoffice.org. However, the beauty of open source is that it can be forked by anyone who chooses, as was done today. Our sincerest goal for Open Office is that it become more widely used so if this new foundation will help advance Open Office and the Open Document Format we wish them the best.

Which I interpret to mean that as far as Oracle corporate are concerned, this is double-plus unspeak for the Document Foundation can fork off, without any support or cooperation from the Oracle funded OpenOffice.org project.

Hello Terry,

Yes, it appears that Oracle has decided to speak to this finally.

There was a mail sent to the marketing ML by Martin Hollmichel at Oracle:

issues regarding the Document Foundation should be discussed within that project, not within OpenOffice.org.

I consider such marketing materials as a conflict of interests between two projects with different objectives. I'd appreciate the removal of that material and also would expect the discussion of LibreOffice related issues to happen elsewhere but not here.

It will be interesting to see what happens to the volunteer contributions for OO.org. I would not be surprised to see contributions apart from Oracle employees dry up entirely over the next few months.

I don't mind continuing to maintain the status quo here, because we're helping users and not Oracle directly. Most users will continue to be OO.org users for quite some time, I expect, because of OOo's name recognition and reputation.

However, if I were directly contributing material to OOo, like patches, docs, wiki text, marketing material, I would prefer to contribute directly to TDF/LO, and let Oracle fend for itself. I expect that many volunteer contributors will feel the same--it certainly seems that Oracle has negligible interest in the community, other than to receive contributions from it. Why should I contribute to a "community" project that has only one participant?

Sun at least seemed to value community participation, even if they were not exactly adept at building it.

We are not an OpenOffice.org project. IMHO, I don't think it's our job to take partisan views here, nor to be antagonistic, nor biased to any particular fork. Our purpose is to support the wider OpenOffice user community whatever the flavour of the product they happen to be using: OOo, NeoO, LiO: business as usual.

If Oracle is unhappy with this, then its representative can contact the community directly and we can evaluate and respond based on this content.

It's very nice to see that this forum will help those using Libre Office. The help I received from the experts here made use of Calc really enjoyable.

I think all that details clears the last doubt, that Oracle thinks not only the name 'OpenOffice' but also the program and the community work belongs them - SUN already tried to appropriate them the community work. And as Oracle don't seems to be a filantrofic association (so few as M$), it's also clear that it will not take many time until there isnt no free version of Oo and one have to buy it.

This affirmed a-posteriori that the fork was not only convenient, but necessary. Bring we now all attention and manpower to LibreOffice and forget OpenOffice.

This occurence, but also the happenings with Java and MySQL, are very important to clear also the last doubts that any 'patrocine' like SUN's of open projekts have to be avoided inconditionally, and that nobody should contribute to such projects, specially not without possibilities to licence their work parallel in a second form and/or to reuse it for another open project or fork. Thus, in future, will we observe more rigorously this point not more to contribute to 'patrocined' projects. And all projects currently in this situation, should be forked still today, for don't more any work what later a firma can thieve.

It's right also to advice the users, w.r.t. all 'patrocined' programs, that there is no garany for freedom and continuity in the future, that it's not really free, and that it can happens at any time that for the program users will have to pay.

TerryE wrote:We are not an OpenOffice.org project. IMHO, I don't think it's our job to take partisan views here, nor to be antagonistic, nor biased to any particular fork. Our purpose is to support the wider OpenOffice user community whatever the flavour of the product they happen to be using: OOo, NeoO, LiO: business as usual.

If Oracle is unhappy with this, then its representative can contact the community directly and we can evaluate and respond based on this content.

I agree entirely, and I think the distinction between users and builders is important. The comments from wl59 may be valid for the developers and others driving the progress. By all means, open solutions should be kept open, and to some degree the users need to be made aware of the implications of what's happening.
However, this is a user forum, and the average user has little interest in the finer details here. We do not build any user base by stepping on the barricades for a cause that's not really interesting. Refusing to support proprietary branches, or worse, abandoning this and other established forums because they bear some "OpenOfficexxx" name, would be taken as "support is not reliable", and users at all levels (home and professional) will look elsewhere.

+1, and a very big one, pun intended.