According to this Linuxworld article, OpenOffice.org 2.0 will be delayed until at least June or July because of the level of feedback received on the beta. The project’s other significant problem, according to developers at this year’s OpenOffice.org conference in Canberra, Australia, is the shortage of developers not tied to Sun.
OpenOffice.org 2.0 delayed until July
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
OpenOffice must be a behemoth plus some ugly (cludged together over the years) code.
So, is Sun hiring or do they expect people to work on this stuff for free? IBM could do more too, but I guess it’s too much of a mess for them.
Easier said then done, but it sounds like some person(s) need to attempt to rip the guts out of it, salvage what you can, and go forward.
At its present state, the code looks like its unmanageable.
Has anybody that has looked at the code want to comment?
I wouldn’t be surprised if they developed a word processor for Eclipse at some point down the road.
Technically, Eclipse is already a word processor. At least on the same level as a typewriter.
Plus, you can code HTML with it, which is like being a word processor.
I’ve always wondered what would happen if Sun decides Openoffice is not worth their time and stop work on it. We’d all like to believe OO.o would survive, through the magic of open source, but would it really go anywhere? Of course the code would still be around, but a project of that size quickly becomes irrelevant if there are no serious, dedicated developers involved in it.
Sounds like Sun has failed to drum up the developer support for the project that they had been hoping for. It just doesn’t sound very enticing to hack away at a giant hunk of code, which by all reports, is quite convoluted in parts.
I’m surprised by the delay! Openoffice beta 2 is really good, it comes closer to the MS office interms of ease of use. Yet its far behind in parsing the MS doc formats.
The odt format really rocks, hope it will make major impact on market. The speed of starting the Openoffice is still very slow. Looking at the free office kit its really worth downloading and using. Honsetly how many of the featrures we use? I guess 10% of the total features… so one needs to find out what best suites his requirements. I wont say everyone should go for open office. But nothing bad in trying?! If it sutits your need go for it else MS is always ready to suck money from your pocket!
Hope the ver 2 will be out soon. I’m pretty excited for the version 2.
This is why projects like KOffice exist. They’re still small and uncluttered enough that the barrier to entry is low enough for normal developers to get into in their free time. Perhaps KOffice will never make it to the level of features as OpenOffice, but it is very valuable as an “Office Light”. If I want to type up a letter or report with some graphics and light spreadsheeting, which is the majority of what I and many others do with an office suite, KOffice is good enough, and much faster. Same goes for abiword and gnumeric.
I hope there’s a decent Word Count feature on 2.0 final. How long will they keep delaying this? And no, the one on the last beta isn’t good enough.
Geee, for such a long time, I’ve been hearing all ’em GPL zealots whine over Sun and how awful they are… now suddenly everything of the success for Linux desktop lies in Suns hands. Then it’s possible to come back whining?
Geee I wonder… why isn’t IBM doing anything? Maybe because they couldn’t care less…
This is the one thing I do like with OpenSource – the willingess to say, “no” to releaseing a product that isn’t up to scratch; maybe commercial software companies should take this onboard as an example of how to do business.
I totally agree with you. No use in rushing something to the front without checking things over and over (ask the Germans and their first batches of Panthers and Tigers in those nice 40’s). The stuff you release should be without clear bugs.
OO.o does look fine indeed. Also feels kind of good and responsive, but must admit that I checked it on a fast x86 machine. So I don’t know how fast it actually is.
People that complain about this setback should either:
– Donate: http://www.openoffice.org/contributing/donate.html
– Participate: http://www.openoffice.org/contributing.html
– Or get one of those nice alternative packages around: http://www.amazon.com/exec/obidos/tg/detail/-/B0000AZJVC/qid=111398…
This is a good thing, they received so much feedback they want to include in the release, and as such, delays the release? This is the best decision they could make. The more impact v2 makes the better – all eyes are on OO.Org – be very sure many analysts and industry insiders will be watching the release, because its such a significant threat to the Microsoft Office monopoly, and the OASIS standard is really something that should be used in government offices and industry alike, because it doesent force you to have windows+office just to read a damn document.
Kudos to the OO.o developers
Geee, for such a long time, I’ve been hearing all ’em GPL zealots whine over Sun and how awful they are… now suddenly everything of the success for Linux desktop lies in Suns hands. Then it’s possible to come back whining?
——
The reason that openoffice.org is even buildable xternal to SUN is because of the presence of Ximian OObuild
http://ooo.ximian.com/
SUN requires everyone to sign an agreement which gives them exclusive access to use the OO.o code within a proprietary product which explains why companies like IBM dont want to get themselves involved in this project
For a better to manage these kind of projects take a look at the independant Eclipse foundation
OOo isn’t at all late, if you ask me. It’ll be delivered exactly when it’s finished, and that is the time when any program should be delivered.
“Around theese times” it should have been delivered, they say… June/July _is_ around now – it’s only three months to July, isn’t it?
Look at Longhorn; now there’s delays for you!
OO.o behemoth is in getting a new whonowwhatsit toolkit working on all the platforms. That is compile time plus the millions of lines of code.
My best wish for OO.o and Sun (God bless them, by the way…) is for Sun to license a cross platform toolkit (my suggestion is strongly for QT, maybe version 4) and work the way out for the toolkit.
This would allow, a) to low compile times down, b) to have a more consistent look and feel on most of desktops, c) to lure developers on contributing
The drawback would be for Sun to define licensign terms with TrollTech allowing developers to work on the windows version of the suite.
Or to buy Trolltech altogether…;-)
Hi all!
Does anybody know what happend to OO.o? Since version 1.9.91 the sturtup speed has really reduced
(I’m using Linux version if it makes any differance.)
it would really be nice to see sun endorce and use Qt (especialy 4) but they are GNOME fans and i dont think they are going to change that. though i must say I hope they do. I would love to see Solaris running run a nice new Qt 4 built desktop and OO use Qt 4
I totally agree with you. No use in rushing something to the front without checking things over and over (ask the Germans and their first batches of Panthers and Tigers in those nice 40’s). The stuff you release should be without clear bugs.
True. I think most of us can accept that not all bugs can be fixed, but what they should do atleast is:
1) Write down ALL the features that HAVE to be included in the next version.
2) Work their ass off to get EVERY one of those features into it – even if it mean going 3 weeks over the release date.
3) Issue a beta, and ask that EVERY beta tester to submit bug reports, no matter how small.
4) After 2 weeks of open testing, close the bugzilla to any new submissions – spend a week prioritising those bugs – some are feature requests hidden as bugs, whilst there are bugs which are as a result of crappy configuration on the end users part (distribution related issues).
5) Fix ALL those issues, rebuild the tree and issue another release.
6) Repeat 5 and 6 until the number of bugs reach zero – in other words, one would have to be INCREDIBLY unlucky for OpenOffice.org to crash by the time it is released – it will not be 100% bug free, but it will be very close.
You read my mind regarding Qt. Best thing SUN could do would be to fully buy out Trolltech, and BSD the *WHOLE* Qt library and simply charge for development tools and consultancy – and for the C coders, create a C wrapper for Qt (if possible) as well as Java.
The great thing with KDE, all the heavy work in regards to a messaging system etc. is all done, all that needs to be done is clean up the UI, and it would be a perfect desktop solution.
Ah, but SUN would rather sit on the $7billion and rant about how they’re going to create new “distruptive licensing schemes” – the only thing I see distruptive is their crappy share performance and pulling the NASDAQ down further.
it would really be nice to see sun endorce and use Qt (especialy 4) but they are GNOME fans and i dont think they are going to change that. though i must say I hope they do. I would love to see Solaris running run a nice new Qt 4 built desktop and OO use Qt 4
Qt and Gnome don’t contradict.
“Sun Microsystems’ chief technology evangelist Simon Phipps acknowledged the challenges OpenOffice.org faces and put it down to its monolithic code base rather than Sun’s contribution governance.
“For something that was originally written for Windows 3.1 and OS/2, the fact that it now runs on Linux and Solaris is a significant achievement,” Phipps said.”
As if Sun had anything to do with the fact that this code runs under Linux…..
Sun did open source the original StarOffice code. Since then people have vastly improved it, much of it has probably even been re-written. But All of sun’s contributions to OpenOffice pale in comparison to what Star Division originally invested in that code.
As much as I like OpenOffice I don’t see a real bright future for the product. Don’t get me wrong-I don’t see it vanishing or being abandoned-but I don’t see a lot lof likelihood of break through innovation happening. Some of this is undoubtedly due to its affiliation with Sun. Sun does not have any history of successfuly building development communities on the open source world. Much of it is probably due to the fact that this beast has already been written-the challenge to create excellent new software is mitigated to a large extent by the fact OpenOffice does most everything and does things in a very spercific way-leaving relatively little room to try out new things implemented in a different fashion. If I was a developer I would far rather invest my time and energy in Abiword/Gnumeric or KOffice
The article mentions 50 developers in Germany, I wonder how many of these are the same folks from the original Star Division. Sun has had 5 years now to help facilitate the building of a community-and their responsibility for this is directly tied to the honerous requirements that they impose on potential code contributors. If these requirements were not an issue it would simply be a matter of whether or not there is sufficient community interest to further drive development. But as it stands it’s hard to gauge potential community interest due to many simply not wanting to work for Sun….
Has anybody that has looked at the code want to comment?
The OO codebase is absolutely god-awful, and it is nigh-on impossible for anyone other than someone with the commitment willing to work on it full-time to do anything with. I just don’t think it’s sustainable. OO is not a project where you’re going to get lots of little patches from part-time contributors. That UNO code especially, oh my God. Also, if you want to contribute code directly then you have to sign one of Sun’s documents to do it, although fair enough, they’ve got their reasons. The only other way is through Ximian’s (now Novell) build process, but in the eyes of the OO community you’re anonymous if you do. I don’t think many people contribute through that way anyway, other than Novell employees.
Just have a look at the Ximian hacking document on Open Office. It has the words ‘crack smoking’ liberally scattered around the whole thing. I’m amazed that Open Office is as feature complete as it is.
i downloaded open office 2, i get the tar.gz file, i uncompressed that and i got some rpm files, wtf?
quite disappointed, i used rpm2targz
I personally prefer an office suite that went through decent Q&A tests than something that will have a brown paper bag release the very next day. What’s the rush anyway? Is MS releasing new MS Office in June?
My best wish for OO.o and Sun (God bless them, by the way…) is for Sun to license a cross platform toolkit (my suggestion is strongly for QT, maybe version 4) and work the way out for the toolkit.
A full-featured office suite written with Qt 4 would be extremely interesting. KOffice 2.0 and onwards should be interesting. However, even then you’ve still got to have the time, resources and developers to do it. However, it would still be better than the mess that is Open Office at the moment.
The drawback would be for Sun to define licensign terms with TrollTech allowing developers to work on the windows version of the suite.
Not really. It would still cost Sun a heck of a lot less to use a decent toolkit than what they’ve got now.
…but they are GNOME fans and i dont think they are going to change that. though i must say I hope they do.
Well, Open Office is hardly a Gnome application.
Best thing SUN could do would be to fully buy out Trolltech, and BSD the *WHOLE* Qt library and simply charge for development tools and consultancy
If Trolltech is ever bought out Qt, automatically goes under a BSD license. However, the only reason people talk about this constantly is because they want to develop something for nothing. You simply don’t make enough from consultancy or anything else to do that, and the development tools go hand-in-hand with the toolkit. The reason why Qt is as good as it is is because it has a funded, working and stable business model. The worst thing that could happen is for that to be upset.
Let me state it clear: i am quite happy with Sun (incidentally, i am J2EE Developer and JCP member), and I don’t want to ignite any stupid KDE/Gnome or QT/GTK debate.
My point is that, actually, QT is cleaner as multi platform toolkit than GTK. BTW, I am talking about toolkits, not desktop environments.
I know KOffice, appreciate it, but integrating in a Windows centric world is better done trough OpenOffice. I am also KDE fan (3.4 on sid at the moment).
I also want to say that i think TrollTech is doing lot of good work, and the joke about Sun buying them for me is right that, a joke.
The point for the whole thing is making OpenOffice development faster and more open to external developers.
Currently it goes this way, more or less : “hey, the word count feature sucks, let me fix it”, fix it, then hit make, and the build take about 10 hrs on a current x86 machine, even more than XFree or Xorg (other monolithic monsters…).
Don ‘t know even if the beast compiles correctly under distcc, so…
The line about Sun licensing etc was about Sun getting licensing terms from Trolltech allowing the development of OpenOffice with QT free of charge for contributing developers.
I know it’s all a dream, but, hey, five years ago i was dreaming of a windows free machine, and now I am writing this on my Debian….
Could anybody forward this to Sun Microsystems? I would also apply for Open Source Vice President position, if still vacant…
QT isn’t going to save them. The codebase is 10 million lines. As the article mentioned, there’s cruft in there from 20 years ago, and it’s a big monolithic mess. It’s way beyond slapping another user interface is going to solve something.
Why buy Trolltech when there is a good cross-platform open source C++ library that has lots of features and is licensed under LGPL ? see http://www.wxwidgets.org/
Under Windows and MacOS it uses the native controls, under Linux it uses GTK or X11.
And it’s not just a GUI library, it supports networking, disk I/O, multi-threading and more!
The current pre-release of OpenOffice is fragile in some areas. Better to let folks chew on it and get it right for the next major release.
(Yes, I’ve filed defect reports.)
Full xorg rebuild takes 30 minutes on Athlon XP 1800 + 512 MB RAM.
> Why buy Trolltech when there is a good cross-platform
> open source C++ library that has lots of features and is
> licensed under LGPL ? see http://www.wxwidgets.org/
Because Qt is *much* better and mature? Because it has already been *proven* that you can write a great desktop with it?
> If Trolltech is ever bought out Qt, automatically goes under a BSD license.
No. The way you put it this it not true. I don’t know why this misconception is repeated in so many places.
Please read the rules laid down in the Free Qt Foundation. *If* the buyer (or Trolltech, no need for a takeover) fails to release new versions of Qt under the GPL for too long *then* the foundation can decide to release the last GPLed version under a BSD-style licencse.
See: http://www.kde.org/whatiskde/kdefreeqtfoundation.php
I have been using OpenOffice Beta 2 on Windows and FreeBSD. I have used the Writer, Calc, and Impress applications somewhat extensively and have filed a number of non-anonymous crash reports with Sun. Of the three applications Impress needs the most QA work.
So OOo got delayed 3 months, what’s the big deal? It should be delivered when is done according to their QA goals and not based on artificial date goals.
Why buy Trolltech when there is a good cross-platform open source C++ library that has lots of features and is licensed under LGPL ? see )
I know a few developers who were willing to contribute to OO.o but didn’t because SUN wanted them to sign some silly contract or agreement. If anybody is killing OO.o it is SUN.
I like OOo a lot.
and for all the neigh sayers out there:
At my previous job, I used it on a Windows95 OSR2 box on a Pentium 233mmz with 96MB RAM. It takes so long to start up, that I simply turn off the monitor when not using it.
It was on a computer used primarily with a piece of $15,000 audio precision equipment, however I needed a spreadsheet program to convert the ASCII data to charts that are easy to compare against other charts in a format that is easily distributable.
Once loaded, it ran very well.
PS, the message boards at OpenOffice.org are extremely helpful.
If they want to take 1 full year to make the FREE software better, I’m all for it–as long as Betas are released publicly along the way, WHO CARES?
I wish em all the best and will continue to post any bugs I find on their support forum.
Note also, that KOffice needs support too and may turn out to be the better option for linux users.
Let me get this straight, OpenOffice’s major release slips by three months and everyone is hysterical. Meanwhile NT 5.0 slips a full two years, XP slips another 2 years and Longhorn slips at least 3 years and no one claims that MS is on the verge of collapse. Give us a break guys. Whether made public or not, slips of three months or more are routine in the software industry. Often times to make a public release date features are paired back (a la W2K) in order to make even modified release schedules. So lighten up on these guys.
On the code base issue. I was under the impression that OpenOffice 2.0 was a ground up rearchitecture of the code base. Was that just at the GUI level and the lower level code is still a mess?
OOo 2.0 is not a rewrite or a rearchitecture or anything like that, period.
I think you are referring to VCLplug? VCL, Visual Class Library, is the cross platform toolkit layer OOo is using. VCLplug allows OOo to “plug” native look-and-feel, instead of system-independentant one. OOo 2.0 will have VCLplug for Windows XP and GNOME. KDE situation is unclear.
If they want people to participate they must refactor the whole thing to allow building only certain parts(separete the core libraries, the word processor parts, the spreadsheet, etc). It has no sense fail to compile due to problems wiht the spreadsheet if you want to contribute to the word processor.
I as I understand it the IBM editors for Workplace 2 (built on the Eclipse platform) are an IBM fork of the OOo codebase. How it is licensed I don’t know but I don’t think it is open. I suspect that it is some closed binary only distribution using an agreement under the SISSL sith Sun.
It would be good if IBM opened this as a plugin for Eclipse.
> Note also, that KOffice needs support too and may turn
> out to be the better option for linux users.
And for Windows users, too, as soon as KDElibs etc. are ported to Windows. 😉
No, seriously, I don’t know how that port will turn out if it is ever really done. I’d really like to see a comparison of the startup time of KOffice/Win vs. the one of OO.o v2 if someone’s really going to do it…
I keep looking at that word “beta” and thinking “user feedback. They must be listening and making changes.
umbergh wrote:
[i]QT isn’t going to save them. The codebase is 10 million lines. As the article mentioned, there’s cruft in there from 20 years ago, and it’s a big monolithic mess. It’s way beyond slapping another user interface is going to solve something.[i]
Perhaps. But stripping down all the UI code plus all the other stuff (think clipboard, printing, java bindings…) could begin to lower the hurdles.
I think this could, or even, should be a target for OO 3 (or 4). Most OS projects rely a lot on corporate developers, ie, developers working on OS projects in office time, but no one of them can survive without the single programmer.
If I’m understanding you correctly, what you’re looking at is the installation files (a la installShield’s “data.cab”). Run the “setup” program and OOo will install itself to whichever location you choose.
On a multi-user system, I recommend installing to /usr/local/OpenOffice-[version], then creating a symlink to that (as /usr/local/OpenOffice). For each user, you can then run /usr/local/OpenOffice/program/setup and install a workstation version (2.2MB per user).
For an upgrade, just install it to /usr/local/OpenOffice-[new_version], change the symlink, then modify each user’s ~/.sversionrc file to contain the new version number. (I’m hoping they removed this step in the 2.0 betas)
KDE situation is unclear.
The KDE situation is not at all unclear. Have a look at Suse 9.3 or the NLD. Open Office provides full support for KDE look and feel.
I’ve always wondered what would happen if Sun decides Openoffice is not worth their time and stop work on it. We’d all like to believe OO.o would survive, through the magic of open source, but would it really go anywhere? Of course the code would still be around, but a project of that size quickly becomes irrelevant if there are no serious, dedicated developers involved in it.
You hit the nail in the head. Remember what happened just now with FireFox. It is a huge project on the spotlight, and still, it only has 5 base developers. And the core developer of it compained a month ago that the other 4 are not very active as of lately.
“OOo isn’t at all late, if you ask me. It’ll be delivered exactly when it’s finished, and that is the time when any program should be delivered.”
By that standard, Duke Nukem (TM) is just in time… 😉
Does anybody know what happend to OO.o? Since version 1.9.91 the sturtup speed has really reduced
(I’m using Linux version if it makes any differance.)
Version 1.9.91 worked pretty well here. I had one gripe though: no support for KDE in precompiled binaries – only the native oo.o interface. Which is fine (I’m just curious how it looks like using KDE native widgets). In fact, it is more than fine, and I’m grateful to the openoffice team @ freebsd for providing up-to-date binaries for all localized versions (hungarian included) as well. Since 1.9.94 entered ports just yesterday (and there are no binaries yet), I’m in the process of building it right now. I tried to build 91 as well, but ran out of space. Which leads me to my question: I’m sure that I haven’t enabled debug support, yet the work directory grew to 6Gig until I had to cancel the build. What might be the cause of that? Also, build process seemed to be much much slower than 1.1.4. I tried to build it with the following make knobs: -DWITH_KDE -DWITHOUT_MOZILLA -DWITHOUT_JAVA WITH_TTF_BYTECODE_ENABLED=yes. Currently I’m trying to build it without disabling JAVA to see what happens – and to see what’s up with the startup speed (which version exhibited the slowdown?).
KDE situation is unclear.
The KDE situation is not at all unclear. Have a look at Suse 9.3 or the NLD. Open Office provides full support for KDE look and feel.
Just ./configure with –enable-kde.
There’s also a web page: http://kde.openoffice.org .
BTW, the beauty of the SUSE approach is that both KDE and GNOME/Ximian integration is provided out of the box: OpenOffice.org automatically has the right look-and-feel, according to the desktop environment one is in.
Besides OpenOffice.org 2.0, KOffice (with its many integrated modules) is also very promising, IMHO…
This may sadden some consumers but at the same time expected. Reason I say that is because like most developers whether open source or commercial tend to delay a project final release date if bugs need to be resolved or consumers needs do. I’d rather wait a few months for a fully functional piece of software than have one with bugs, etc.
The problem with using QT is that is QGPL or GPL (as of v.4), so you either buy it (and can have closed source) or use the GPL version which means that your whole program must be GPL – not something SUN would like.
“It’s 10 million lines of code and takes serious commitment just to compile the thing,” Foskey said.
OMG! Really! welcome to big time development process! People don’t want to work on a Cluster F*** of code and not get paid for it imagine that! All this and still no the quality of MS office. Proves my tired point of there is no such thing as something for nothing. Where incentive doesn’t exist ultimately a product doesn’t exist.
Unix has always been the perfect fit for opensource because Unix has always been a million little pieces interlocked together to give you a system. This has always meant that enhancements and fixes can be isolated and fixed quickly. It also has meant that project scope was reasonable and people could easily enter a project when someone else leaves.
The biggest hurdle open source faces is having a monolithic monster like openoffice with a mish mash of code and extremely poor design. What the community needs to do is a) if they are able to wrestle control away from Sun, fix the design of Open Office by making it very modular or b) start a new project and build the tools right so that it can flourish in an open source, decentralized development style.
I think that has more to do with the licensing arrangements rather than source code: the dual-licensing program with SISSL would discourage many of Sun’s potential competitors from participating in the main project, which shoots down a good reason to use OOo (if, say, IBM forks OOo, they have little to gain because they can’t collaborate with other OOo developers).
Sun seems to want the best of both words (open source collaboration and propeitrary commercialization) without realizing it doesn’t work. Their efforts to make sure thy are able to use OOo to make StarOffice while preventing any other contributor do the same clearly backfires on them.
Kim: With the GPL edition, it allows you to use GPL-compatible licenses, amongst which is the BSD license. OpenOffice.org can be in BSD license while Sun can still sell StarOffice with a license from Qt quite propeitarily.
amen 😉
the problem with it is more people will appear crying about dependency hell and too much libraries to compile… And Open Office is condemned to be monolithic and WYSIWYG like MS Office or to be refused
angustia: the problem with it is more people will appear crying about dependency hell and too much libraries to compile… And Open Office is condemned to be monolithic and WYSIWYG like MS Office or to be refused
You can always distribute it as a single unit even if the system itself is modular. That’s what I’ve done with my projects.
Make them modular for ease of programming, compiling, debugging, and profiling. Then when distribution time comes, package it all together so it looks like a single unit and ship it off.
OOo code is horrible to work with. Remember the Openoffice bounties?
http://www.gnome.org/bounties/OpenOffice.org.html
I tried doing one of the simpler ones (Image save). I spent a whole day just trying to understand the file structure of the source code, let alone understand it. It took me a week to implement the feature, which worked well.
And then hell broke loose over a menu item. No matter what I did, I couldn’t add a frigging menu item to the context menu. After a week of trying, posting to forums, talking to people on IRC I gave up.
The sad part was 95% of it was done but no one could tell me the proper way to add the menu item that would have completed the job. Oh well, somebody else did.
I use Abiword.
>>This is why projects like KOffice exist. They’re still small and uncluttered enough that the barrier to entry is low enough for normal developers to get into in their free time.
and Abiword.
doh you mentioned abiword. sorry.
>>If they want to take 1 full year to make the FREE software better, I’m all for it–as long as Betas are released publicly along the way, WHO CARES?
You miss the point. The issue here is an apparent shortage of non-Sun coders. I’ve heard the arguement before that the problem with open source is there’s more takers than givers and on the surface this looks like a true example of that.
Some here have (correctly IMO) pointed out that the reason is that OOo is too monolithic to expect a lot of open source coders to be able to actually participate enough. Whereas those with some time can actually contribute with more of an impact to abiword or koffice.
which version exhibited the slowdown?
Sorry for misleading you. I was still sleepy when I was writing my previous post. I wanted to write that the startup time had really decreased. Now work with with OO.o is a pure pleasure (OK, maybe something still crashes, but you only need to press ‘Send’ button and in the next version the problem is fixed).
I share your sentiments (althoug I still use OO.org). While some people have said the Sun copyright contract is a problem, I don’t think it is. It’s just a fact that 10 million LOC is really really hard to work with.
I also looked into one of the bounties, but put it aside after realizing I don’t have the personal resources to do it. Even with a broadband connection, it took a half a day to download everything from CVS. Configuring wasn’t that big of a deal, but on my old PC, it took well over a day to compile the whole tree. The edit-compile-test cycle is also slow, because the source tree is huge. When I attempted to compile for debugging (–enable-symbols, IIRC), I ran out of hard drive space (the build literally takes gigabytes of space)!
The moral is that to develop on OO.org, a person really needs the fastest CPU/disk combination they can buy, probably dedicating a whole SCSI disk just to the source tree. I could see SMP being necessary, too.
Will Fedora Core 4 be released with a beta or will they roll back to OO.org 1.1? I know the test versions contain the beta thus far.
> The problem with using QT is that is QGPL or
> GPL (as of v.4),
Are you talking about the windows version here? Qt/X11 has been GPLed for years.
In addition to the QPL (not QGPL) and GPL there is the proprietary license you talk below.
> so you either buy it (and can have closed source) or use
> the GPL version which means that your whole program must
> be GPL – not something SUN would like.
Are you really saying that SUN cannot afford 60 x 1500 € for the complete developer staff (probably even less because of the number they’d buy), as there are no royalties or per-user licenses for Qt. They’d even get excellent support.
If a company the size of SUN cannot even shell out < 90000 € they must already be bankrupt. It seems I have to call my broker immediately to sell all SUN stock! No wait, that would be insider trading.
If you mean monetary incentive… may be. But most coders and hackers volunteers on Open Source projects aren’t doing it for the money. Personally, I don’t see the 10+ million LOC being cleaned up overnight by some hyperactive volunteer staying up every night for the next year or so… but may be if somebody like an angel investor stepped in… perhaps African entrepreneur Mark Shuttleworth 🙂
I’m using v1.9.91 at college and it’s mighty good. No major problems importing MS files, yet.
I don’t know why some people even think about that IBM could get involved in OOo? IBM has a finished Office suite since they brought Lotus (Sotus SmartSuite) at it’s time itwas pretty good, but IBM didn’t continue development of it, after the port to OS2 failed quite misreably. So why should the get involved in this Sun-controlled project? Besides I wish they would refresh SmartSuite with AmiPro and Lotus 123. These were really good products.
“I don’t know why some people even think about that IBM could get involved in OOo? IBM has a finished Office suite since they brought Lotus (Sotus SmartSuite) at it’s time itwas pretty good, but IBM didn’t continue development of it, after the port to OS2 failed quite misreably. So why should the get involved in this Sun-controlled project? Besides I wish they would refresh SmartSuite with AmiPro and Lotus 123. These were really good products.”
Wrong as I pointed out before, the IBM Editors for Workplace 2 is an IBM fork of OOo embedding it in an Eclipse based platform.
http://www.eweek.com/article2/0,1759,1592319,00.asp
Download and take a look at the pdf here:
http://www-1.ibm.com/support/docview.wss?rs=1097&context=SSLJ3H&dc=…
IBM has given up on SmartSuite and IBM is already “involved in OOo”.
“Besides I wish they would refresh SmartSuite with AmiPro and Lotus 123. These were really good products.”
SmartSuite was _far_ better than MS Office. I really enjoyed using AmiPro way back–it was so much easier to use than the kludge that was MS Word was. Quite frankly, MS Word was a real POS (arguably still is).
On the other hand, Sun, Novell, and Red Hat all are backing OO.org in their uprising against Microsoft. OO.org has basically become the defacto UNIX/Linux desktop office suite, which IBM would have to face in their own Linux efforts.
While OO.org lacks the intuitive interfaces of SmartSuite, it is definitely a shade better than MS Office, once the user gets aclimated. There are certainly things about OO.org that seem odd at first, such as having to create new document-wide colors before using them in a document, but it isn’t terrible.
“Wrong as I pointed out before, the IBM Editors for Workplace 2 is an IBM fork of OOo embedding it in an Eclipse based platform.”
Has IBM submitted a copyright agreement with Sun, so their patches can be put into the main-line source tree? If not, then IBM really is playing the not-invented-here game.
From the eWeek link: “We will not advertise these as OpenOffice editors; we will be clear that they are OpenOffice-derived…”
I guess I answered my own question!
Agreed. The ridiculous thing is that IBM will not publish the file formats for Lotus so taht the likes of Abi and OO.org can open these files.
Why the intuitive itnerface has not been taken on by more open source developers as an excellent model to follow, I do not know. It set a superb benchmark and would have complemented the GNOME HIG, for example. This has been raised amongst the GNOME developers more than once.