When I first joined the KDE party, I was using Blackbox (and eventually due to my KDE involvement got to meet Brad Hughes, author of Blackbox :) and the only KDE app I used was KMail. KDE hacking was at first just a fun way to pass the time during some lull moments in my life. That seems like .. quite a while ago now. :) Even back then, KOffice was around.
When it started there was no other F/OSS office suite. Today, there is OpenOffice which "rules the roost" and then some office document apps like AbiWord (which is doing some rather cool things on the collaboration front). KOffice is the only other "suite", however, by which I mean it is a collection of applications that shares a strong identity through shared technology and design.
Even though KOffice 2 was recently released and has some really great ideas at its heart, it still plays to a bit role part in things. This is pretty unusual for a set of KDE applications that has been around as long as KOffice has been. Usually KDE applications post really good to very strong showings among F/OSS users. What has held it back? A few things, prime amongst them high fidelity support for Microsoft Office formats and a lack of effective promotion.
KOffice has always had a relatively small team, but it's been a gutsy and productive one. The team has been growing lately, though. There is a consulting company doing KOffice work now and they are slowly building up paid developers. Nokia has been working with pieces of KOffice for a document viewer for Maemo, which is great news for the MS Office compatibility front. KOffice people recently had a successful developer sprint and have moved to a use case centric development strategy. Even the icons are turning all Oxygen glory thanks to Pinheiro and the Oxygen Icon Cohortory. (yes, I just made that word up. ;) So things are growing and KOffice is on its best footing since I've been around.
I also think that the promotion of KOffice is getting aligned nicely. First off, KOffice is more than just a set of word, presentation and spreadsheet apps. It's also Krita, which is turning into one of the best image editors in the F/OSS market, and Kexi, a great database application creator. Both are very solid tools, and Kexi helped blaze trails for KDE on MS Windows. Other apps like KPlato for project management also exist. The promotion of KOffice as a development library for these kinds of applications plus a nice set of "not just office docs" apps is starting to come around.
Ok, fair enough. So why did KOffice make my list of Key Quests for 2010? I think 2010 is going to be a pivotal year, one way or the other, for KOffice and if it goes well it can rise the tide for all of KDE.
If KOffice can plant a flag on Maemo and other similar platforms, even if it is "just" the document display parts (and not the full word processing or other apps in all their UI splendor), this will be an important step forward for KDE in the device spectrum effort.
If KOffice can stabilize the key apps and continue to grow the contribution community around it, the lightweight, integrated, featurful and "good times UI" feel they have developed could propel them into a contender alongside OpenOffice. This is so critical because OpenOffice is not the most healthy of projects, and hasn't been for some time. It's not dead nor is it dieing (nor do I want it to), but it isn't exactly springing forth with life and the code base isn't exactly the most manageable thing in the world either. F/OSS would be much better off with a nimble, highly useful set of Office apps that can play back-up to (or even first fiddle eventually with) OpenOffice. It gives us more eggs in our basket and fewer single points of failure.
For KDE, it would give us a very compelling increase in our proposal to the world if we have a good office suite (and all those other crunchy happy apps, too) that (eventually could) have hooks into things like Nepomuk and Akonadi (share-what?), that start up quickly and which get the job done.
By the end of this year, I think we'll have a very clear indication of whether KOffice will be on its way to that summit or whether it will be another 10 years like the last 10. I like the energy in the project right now and the way they are going about things, and given what's at stake here, it easily made my list this year.
(This article is part of the "Key Quests for KDE in 2010" series)
Subscribe to:
Post Comments (Atom)
9 comments:
IMHO the compatibility to MS and OO issue is crucial for success and day to day usage.
Given the permanent need to exchange documents (at least for office users) there is unfortunately no point to use incompatible apps even if these are excellent as koffice2 is.
yes, for apps like KWord and KPresenter. having OpenDoc for their native file format is a big plus, and the huge amounts of work happening right now in the MS Office filters as well as OpenDoc support is critical (and i allude to these efforts in the blog entry).
that said, MS Office / OO.o compat for apps like Kexi is completely irrelevant.
this is where the promotion of KOffice apps needs to step up (and i think it will): to create public awareness that KOffice is not just word-spread-present and that several of its apps live in realms where file compat with MS Office isn't overly interesting or at the very least not a blocker.
that some people will toss aside all of KOffice because KWord doesn't import MS Word files perfectly (though i think that is set to change dramatically in 2010) is unfortunate and a promo failure.
Every App I use is a KDE app. Except for an office suite. Only for the MS filter. Do not get me wrong. I hate MS as much (And some times more) as the next guy. But I need a filter to at least convert a doc, which someone gave me, to odf. I was delighted when I heard the news that Nokia has started work on MS filters.
I think your analysis is spot on, with one exception: why KOffice has been held back as it were.
What I think is the main reason is a combination of a couple of things: Even if KOffice 1.x was ok, it was never really very good. And worse, the code base was a dead end. It was clear and easy to work with, but didn't really allow many new concepts like the docker one that we have now. Also, the compatibility with other office applications was very low due to it having its own file formats and filters that were ranging from outright bad to so-so.
So the KOffice team took on the really large task of moving from our own file formats to OpenDocument. And just when that started to shape up, we had to take on the next task: to move from KDE3/Qt3 to KDE4/Qt4. At the same time we wanted to get rid of the old limitations and build a completely new application framework that would allow us to move the suite into the next decade just like KDE itself did.
Unfortunately, it turned out, that we started out the second task too early and the development was really held back a lot by Qt4 bugs and ever changing API's in kdelibs. So that task took longer than it would have if we had started, say, 6 months later.
The end result of those two things taken together was that there was a large gap of almost 3 years from the last release of 1.x (1.6.3) and the release of 2.0. And after that it was still not ready for end users, which we clearly said in our release announcements.
I think that if we had marketed KOffice harder, we would have a large backlash. What we have now is a very good foundation, and applications on top of that that are ok'ish. The development is so much easier, and happens so much faster than it does in OOo, so I guess that feature wise we are going to catch up in between 1 and 2 years.
And don't forget: OOo has its own toolkits and foundation classes. We get all that for free by using Qt and kdelibs. If we count the people that are working on those two components as well, I am willing to bet good money that there are more people working on KOffice than there are on OOo. So the future is indeed ours.
KOffice is "different". It always has been, and as far as I can see, always will be. That is its strength and its weakness. Apart from the MS compatibility, which others have addressed, the other things desperately required are documentation and training. I'd like to make a proposal to solve both problems.
Forum.kde.org has Klassroom courses. If a KOffice developer or devotee would be willing to co-mentor a course for one of the applications we could teach usage and at the same time get documentation written on userbase that comes directly from user experience. It requires that some time is set aside during the class week, to answer questions and criticise output if documentation is being produced at the same time. It's not terribly demanding, and the result could be really worth while. It has to be worth a try. Volunteers?
I'm a big fan of KOffice, and I just wanted to chime in with another viewpoint here. Not to minimize the issue of document compatibility, because yes, that's important, but I'm really glad that that hasn't been the focus thus far with KOffice 2. I'm thrilled to be able to point to at least one free software "office-y" application and say "See, here's actual innovation! Here is a new thing!"
You rightly mentioned the AbiWord project for the work they've been doing lately, and I think that's true. But overall there's such a follow-the-leader mentality in this space (whether you see that leader as MS Office or OOo), and it's really refreshing to see a project that's not afraid to break new ground, be it in the UI or whatever. I think that work's going to prove to be a really solid base for the KOffice of tomorrow. And if you prove prescient in half of what you wrote here, I can't wait to see it.
Kudos to the KOffice team, and good fortune in 2010!
If I had a million to invest, KOffice would be it...
If I had some more spare time to invest, KSpread could be it...
Apart from that, I'm missing Kivio in KOffice 2 - it's a great app - hope someone's going to port it over.
Go KOffice!
I think an office suite that has true compatibility with MS Office that doesn't mangle documents is the singular most important requirement for Linux to succeed. This will allow people to use Linux and work with people who don't. OpenOffice just doesn't cut it yet; I can't trust that OOo documents come out properly when viewed in MS Office. Based on the project's pace and roadmap, I don't think it's going to get there. I do think KOffice has a real shot.
Koffice!
Aaron, please give a little blessing to the koffice by making the bug fixed:
http://bugs.kde.org/186513
It's been also reported to qt-bugs as upstream, which needs attention...
http://bugreports.qt.nokia.com/browse/QTBUG-7177
KWord which is the nicest application; deserves a little facelift.
here's the ugliless we have to live with :(
http://bugs.kde.org/attachment.cgi?id=34534
http://bugs.kde.org/attachment.cgi?id=31891
Please, if possible, do contribute your energies!
thanks.
Post a Comment