Guest post by Michael Meeks who leads the OpenOffice.org development team within Novell has taken a
detailed look at contributions associated by metrics to OpenOffice.org and makes the case that Sun’s tight control over the codebase and the lack of enough volunteer contributors leaves the development slowly stagnating over a period of time. Michael Meeks has recently started strongly advocating the position that Sun needs to setup a more independent OpenOffice.org foundation or otherwise allow more relaxed policies for commit access and be less rigid about assignment of copyright to itself for the development community of Openoffice.org to thrive beyond Sun developers.
He writes:
A half-hearted open-source strategy (or execution) that is not truly ‘Open’ runs a real risk of capturing the perceived business negatives of Free software: that people can copy your product for free, without capturing many of the advantages: that people help you develop it, and in doing so build a fantastic support and services market you can dominate. It’s certainly possible to cruise along talking about all the marketing advantages of end-user communities, but in the end-game, without a focus on developers, and making OO.o truly fair and fun to contribute to – any amount of spin will not end up selling a dying horse.
OpenOffice.org occupies a prominent position in the Linux Desktop and among Windows users as well looking for a free alternative to Microsoft Office. It is important that it continues to progress further to remain competitive.
At one time, didn’t OpenOffice not have Java as a depends?
BTW, The entire quickstarter should be removed from the codebase. If ever there was a useless class of program, quickstarters would be it.
It’s always been there – at least for the Help system if not more.
True. It just serves the purpose of hiding the long load time the system has always had compared to other products – namely due to its overly object-oriented C++/Java style programming. They could really improve performance a lot, even the latest versions.
Also, it’s pretty important for Ooo to succeed. Many, many individuals I know who cannot afford a $500 Office Suite (I live in the Caribbean) turn to Ooo to help them create their documents etc.
Without Ooo, where would they go for help? If there’s another Open Office Suite with fairly complete programs (equivalents of Word, Excel & PowerPoint), I’m all ears.
Until a competitor or replacement is found (which can run on Windows, I should add), we must make all efforts to save Ooo and make it a better program with every new release.
I suggest that you try Softmaker Office 2008. It’s proprietary, but sub-$100 and has overall better MS compatibility than OOo. Currently it doesn’t have a macro recorder though (it does have a VBA equivalent called BasicMaker).
On Softmaker’s website, you’ll find a 30-day trial version of the software that’ll save to MS formats for 7 days.
Yea but I still run into far too many people that think they need MS Word to open and use a word document. Then if they actually do try OO they don’t give it a fair shake due to the familiarity of MSFT Office. Just like Windows they don’t understand that it’s not that MSFT Office is easier it’s just familiar.
As an outside observer the development does seem slow – Ooo has not yet got the functionality of MS Office 2000, Although for many tasks Ooo is adequate it’s not as good as it should be.
Edited 2008-10-11 06:55 UTC
When there are only two companies actually working full time on it – are you surprised? Novell and Sun are doing as much as they can with the limited resources – its time for the vendors who do bundle it with their respective distributions to actually allocate some man/person power to the project.
Well, there is only one company working on MS Office, too, and there’s only one company working on iWork
. But it looks as if Sun is slowly downscaling their development effort on OOo, and OOo is unlikely to be ever a popular project for volunteers to work on, even if it became truly open and community oriented, because of the size and complexity of the codebase, the weirdness of the buildsystem, the amount of unfamiliar technologies used — technologies that are only used in OOo. But then, as a KOffice developer, I _would_ say that, wouldn’t I
Oh, since I now have a chance to talk to an actual KOffice developer: are there plans to seriously boost MS compatibility, now that MS released their format specs? (Also, you could reuse OOo filters, but it’s possible to do better than that, as far as I can judge by Softmaker Office.)
No. There simply aren’t enough developers around to work on that. We had one student in the google summer of code who worked on improved .doc support — mainly images — but it’s just not feasible to support the MS formats fully, either binary or xml — they are too big, too MS-specific, too much encumbered with things like “do this like Word 95” or “enumeration of copyrighted border decorations”
Besides, with MS promising to start supporting ODF fully before supporting the OOXML “standard”, we don’t have to
Well, at least you’re being honest, but let me translate your comment into what Joe Sixpack is going to take from it:
Um, it’s too hard and boring and we really don’t want to, and you don’t really need that anyway. ODF is better anyhow.
This is not what you said, and I’m not implying that it is what you said, but it is how many people will read it, and unfortunately that’s a bad image to portray.
As for MS supporting ODF fully, I’m on the fence. They could surprise us and actually implement the ODF standard properly, but we could also wind up with ODF an MS ODF, rather like the RTF situation. And what do we do then, I wonder? Unfortunately given MS’s track record of supporting “standards” I’m really not holding my breath for standard ODF. But who knows, there’s a first time for everything.
Well, we’ve always had the same message: see for instance http://dot.kde.org/1194021253. In the end, no matter what the mythical mr. Sixpack may think (if he is capable of cogitation), that’s the situation, and unless someone donates a fairly substantial amount of money to KOffice, it’s where it’ll stay. If you implement a file format, you also need to implement all the features, and preferably in the same way as the application where the file format originated in. For instance, OOo has pretty good compatibilty with MS Word — unless you use sections. This hit me when my daughters had homework for what’s grandly called “ICT lessons” and had to create a document in the MS Word way — but we don’t have MS Word around. In the end, I had to download a copy of Office 2007 with bittorrent to let them finish the assignment.
Thanks, that’s the biggest laugh I’ve had all day. You’ve just exemplified the stereotypical image of most open source software and the developers there of. I don’t disagree with the rest of your comment, but this part…
And just how much did you pay for this copy of Microsoft Office 2007?
Nothing :-). But I also immediately after saving the document for my daughter deleted it. It wasn’t the unpleasant application I remembered — rather better than version 6, which was the last one I used for real, back in the days. Fortunately, I hadn’t completely wiped out the vista partition on my laptop either, otherwise it would have been in vain. Since then, Vista has managed to wipe itself out.
Fortunately, someone at my daughter’s school, a pupil, has started a program to translate all lessons to OpenOffice, so this needn’t happen anymore, and parents aren’t forced to buy Windows and Office for their children anymore.
Give the guy a break. Paying for a bittorrent download wouldn’t have made it any less illegal. Plus you have no idea which country he lives in.
What we’re seeing increasingly in the world now is consumers standing up and refusing to pay the huge prices for software and other digital media/content. I’m not going to discuss the moral issues of this, but eventually the “people power” will win and some things will need to change as a result.
The point this person was making I believe was simply that OOo isn’t 100% compatible with MS Office (yet). Whether he broke the law at any particular time is hardly any of your business…unless of course you work for MS, which I strongly doubt.
There is a test suite for ODF.
I know this is totally foreign thinking to the “One-Microsoft-Way” types, but such a thing actually does exist. Imagine that … standard standards. Impartial verification. Either an application passes, or it doesn’t. Either it has proper ODF support … or it doesn’t really have ODF support at all.
Um, Windows also has support for OpenGL (which is very well specified and also has a compatibility test suite) but look at what a crap driver they produced and barely maintain. They have just enough OpenGL support to get a tick in the box for government contracts that require open standards (in the say way the have a barely adequate ‘POSIX’ support for similar government/military contracts). They obey the letter but not the spirit of supporting open standards.
The ODF support will no doubt consist of very good ODF importers into Office, with an unreliable and underfeatured export. This is the same crap to standard formats they have been doing for years. Yet still people don’t learn and end up losing control over their own data (since they can’t read the formats of their own stuff unless they pay again-and-again to stay on the Microsoft ‘treadmill’). The other real shame is that developers who could have been great are employeed by Microsoft to churn out these turdy implementations, simply because some dinosaur manager can’t stomach the thought of competing on merit alone (heaven forbid customers ever have a choice!).
Don’t say you weren’t warned!
Edited 2008-10-12 22:35 UTC
Thank you for the info! (What I meant mostly was the legacy binary formats, not the OOXML which is not even fully defined, but your answer covers that anyway
)
There are two companies – I don’t know why you turned the fact I said there are two companies into some sort of lynch pin to your whole argument. The fact of the matter is that there are limited resources to not only maintain OpenOffice.org/StarOffice for existing Novell/Sun customers as well as using those same programmers to ad features to OpenOffice.org as well.
Its a whole lot more complex than releasing a piece of software – you have to release it, maintain it, ship updates, test those updates for regressions. Then then there is maintaining those releases from the past. You’ve released 2.4 but there are customers still using 2.3 – so you’ll need to maintain that as well. Again, it isn’t as simple as you make out.
As for OpenOffice.org, there needs to be resources, but like I said, the two major players are doing as much as they can – where is Red Hat? why can’t they allocate 10 full time programmers? or has their ‘giving up on the desktop’ reached as so far as not even helping with desktop oriented opensource products? what about Ubuntu – where is their contribution outside distribution specific bugs?
The analysis is ok. But i am somewhat skeptic that anything positive would come from a Novell employee about Sun or OO.org especially when Novell is involved with MS beyond any other *nix company.
Wonderful, shoot the other side of OO.org development, Linux Foundation guy already shot Sun. I can already see it, Americas crappiest jobs program, Open Source programmer. Shit keeps flying from all sides.
Also i need to add that, Mozilla development is similar to OOo in terms of developer involvement. Problem is code base, not the attitude of companies. if i were them i would write a very powerful Java version that works online like think free..
Mozilla is fortunate enough to receive ~$60 million a year from Google in advertising revenue. Without that much money pouring into Mozilla, they would be in the same boat. I’m sure if companies and individuals invested $60 million in OpenOffice, it would be a much better piece of software.
first off, it’s funny to see how Meeks constantly seems to forget were his go-oo code came originally from… yes, from Sun’s OpenOffice… instead of bashing Sun every now and then, being thankful doesn’t hurt anyone, right?
Of course Meeks wants less Sun branding in OO… Sun competes with Novell… but guess what?… It’s a Sun product!!! why should they remove their branding??? why doesn’t Novell removes their branding from their products?
On the other hand, Meeks has got some balls when attacking an open source contributor as Sun while working at a company like Novell and it’s unclear deals and intentions with Microsoft… look at yourself before judging others, man
Lastly, developers on OO have decreased thanks to the introductions of OO extensions… why doesn’t Meeks takes those numbers in account?… how many extensions have been created and how many people are creating them?… these half-truths aren’t helping Open Source, Mr. Meeks… Perhaps that aligns best to your corporative rulings, hmm?… create confusion and you (MS) will reign…
If Novell is so concerned, why don’t they ask their good friend and business partner for a port of MS Office to SuSE Linux?
–bornagainpenguin
Get a clue before you start bad mouthing Meeks.
You obviously have not been following what he has been blogging about for the past years. If you had you would know that branding is not even *one* of the significant issues that Meeks has raised time and time again. (and copyright attribution is *not* simply a branding issue)
Meeks is one of very few people on earth who actually knows and groks the OpenOffice codebase-and he has been tirelessly working on that code, improving it and making OpenOffice better for everyone involved. The text which he wrote attempts to numerically illustrate the severe problems, which do exist, which he has been calling attention to over and over again, but which has been met with apparently deaf ears by Sun and those managing OO.org.
You may wish to characterize this as some kind of Novell vs. Sun thing. But the fact remains- there are far too few devs working on the project, far too many hurdles to go threw to become a contributing dev, far too many procedural hurdles which actively thwart the good will of those who would like to contribute- and this in the context of a project which is extremely important, not only to the Linux and FOSS communities, but also to a much broader community commited to wresting control of document creation out of the hands of propietary format creators and guaranteeing accessibility to the documents we create.
The viability of the FOSS software in the desktop commercial and government/institutional markets is uniquely dependent upon OpenOffice. If something were to happen to Sun where they could no longer afford to pursue this project and the project would fail due to the lack of knowledgeable contributors-the entire FOSS community would be delivered a profoundly painful blow-OpenOffice is perhaps *the* most relevant alternative to MS Office which exists and makes FOSS desktops even thinkable-without OpenOffice FOSS desktops would not be able to compete at all.
But the stakes are much larger than the relative success of FOSS desktops: what is at stake here is the accessibility of the documents we create, now and in the future, whether the documents we create can be opened and modified using software which is freely available to all and which runs on all of the major platforms, or whether are documents are stored in propietary formats which can only be accessed by propietary software, which is exceedly expensive and which only runs the platforms Microsoft chooses to support.
You bad mouth Meeks and fail to see that his *real* interest in OpenOffice has nothing to do with the relative market success of Novell vs. Sun-his concerns, which he has blogged about quite frequently is the health and longevity of a community of contributors capable of maintaining and improving perhaps one of the most pivotal pieces of software in the FOSS world.
The OpenOffice codebase is horrifically large and complex. Extremely few programmers, given sufficient expertise and capability, have what it takes to become a meaningful contributor. Not only does one need to be extremely skilled to grok the code at hand, not only does it take an inordinate amount of time and energy to get up to speed in the OO codebase, but then one has 10,001 programmatic constraints which makes it very difficult to change something without breaking anything else. Working on OpenOffice is about as unsexy as it gets. And then to top it all off Sun has this extremely heavy-handed procedural system in place, which may have been fine back when the code was not opensource and exclusively Sun employees worked on it, which makes it nigh-impossible for the project to attract enough devs to become a really viable competition.
I just hope that Sun one day begins to understand what community software development is all about and how to encourage, motivate and entice developers to contribute to the project-instead of creating one hurdle after the next and actively discouraging potential contributors. And I hope this day comes before OpenOffice as a project fails due to manpower issues.
OpenOffice, as a project, will never be able to compete with something like the Linux kernel in terms of the number of developers. If for no other reason than the fact that as a new dev that you have remarkably little freedom to innovate in such a godawful complex codebase-pre-existing code which you must accept and deal with, little room to experiment and change, a remarkably thankless task. But given such an initial situation- one must do *everything* possible to lower the barriers for new devs to participate-and this is the simple message that Michael Meeks has been pounding on again and again to deaf ears at Sun and a FOSS community which by and large, although precariously dependent upon the success of OpenOffice, practically ignores it.
Au contraire, I do have been following Meeks for some time now… I even remember last year’s Sun badmouthing (much like this one) and his official/unofficial fork/non-fork go-oo… that had existed for quite some time before that but he re-released as a new/old project for contributors to get their code “upstream” quicker…
yes, I do know what he does… and I do know what OO means for FOSS… but guest what, that IS Sun’s code… and it’s open source, so, whatever could happen to Sun won’t matter, OO is not at any risk… even yet, he chooses to attack Sun, one of the biggest contributor to FOSS ( http://blogs.sun.com/bnitz/entry/is_sun_behind_more_than )..
from my POV, not the best way to promote open source… Novell (via Meeks and others) attacks fellow contributors (like Sun and Red Hat) while joining forces with Microsoft… great job… NOT
You seem to be conflating a few things:
1) Criticizing a procedural system which actively thwarts new contributors and contributions and which makes the already thankless task of hacking on OpenOffice even that much more of a turn off-discouraging developers who might otherwise be inclined to lend a helping hand from even attempting-is not the same as attacking Sun(ie. corporate politics). If IBM was steward of this code and managed it in the same fashion Meeks would probably be lambasting them for their mismanagement.
2) Novell has over 20 years involvement with Microsoft-whether as a company producing software to run under windows, or standing before a judge in legal issues with Microsoft. When I first started working in IT back in 1989 Novell was *the* name in networking for PC’s. They built their business and reputation on a software platform written by Microsoft. Expecting Novell to completely purge any relations to Microsoft after having purchase SuSE and Ximian is like expecting IBM to divest from the mainframe business once they started focusing more on software as a service product.
Moreover I highly doubt that Meeks was consulted by the higher-ups at Novell when they made deals with Microsoft in the past couple of years. You seem to conflate the corporate political issues between Novell and Microsoft with the *agendas* of those employed by Novell who are working on Free Software.
Although I do not have access to any meaningful statistics about who has historically contributed the most to OpenOffice, what I do know is that Sun purchased and already written cross-platform office suite known then as StarOffice. Sun has never been known for their desktop applications-prior to OpenOffice Sun had never produced any meaningful desktop applications. You stated that the code *is* Sun’s code- and this is part of the fundamental problem, regarding the viability of the OpenOffice project.
OpenOffice is FOSS software, at least according to the license, but looking at the community around it(or the relative lack thereof), one might come away with the impression that this is Sun’s baby-which of course is the point you are trying to make. My point, and I believe Meek’s point is that as Free Software OpenOffice is something which belongs to a much larger community, a community which is only limited by those who contribute in one way or another. Sun maintains a propietary hold on this Free Software through a) all the hurdles it creates for contributors b) copyright assignment.
This propietary hold on Free Software is like caging a beautiful bird-what otherwise would be free to fly is held artificially trapped in a cage. Now I understand the commercial needs that Sun has as regards their propietary StarOffice product. I understand that they wish to protect the brandname and quality of that product. But I contend they can still do so without burdening OpenOffice with such a procedural nightmare that devs are avoiding OpenOffice likes the plague.
Looking at the Wikipedia page for Staroffice one finds this rather interesting quote:
” The number one reason why Sun bought StarDivision in 1999 was because, at the time, Sun had something approaching forty-two thousand employees. Pretty much every one of them had to have both a Unix workstation and a Windows laptop. And it was cheaper to go buy a company that could make a Solaris and Linux desktop productivity suite than it was to buy forty-two thousand licenses from Microsoft. (Simon Phipps, Sun, LUGradio podcast.)”
Now why did Sun release the code for OpenOffice under a FOSS license? I don’t really know the answer to this, but pure benevolence it was surely not-more likely they felt they needed to do this to get respect and support from the larger FOSS community-after all Sun was a noname from a consumer PC desktop user perspective, ie. if the project is FOSS it can get uptake by others in the FOSS community and corporations which build on FOSS(like IBM), helping to establish the “street cred” of Sun in the desktop application market. But apparently Sun was not all that interested in getting non sun employees to actually contribute and push through changes to the code, improving it for all.
The structures in place by which means OpenOffice code contributions are managed expresses clearly this lack of desire or interest in 3rd party contribution. And this flies flat in the face of the spirit of FOSS. Xorg, Linux, Mozilla and OpenOffice are all FOSS software. Xorg has blossomed since the old XFree86 days were left behind-Xorg was born out of frustration by devs who were hindered by the XFree86 management to propell X into the 21st century-the license change was only the icing on the cake, the real frustration had already been brewing for a while. Linux has a spectacularly large and diverse community of contributors- who own’s Linux ? those who contribute. Mozilla is struggling with many of the same issues which OpenOffice has-a horrifically complex codebase which scares off many potential contributors, but a) this situation has gotten noticably better in the last years b) they also have spawned a vibrant community of plugin writers-which ultimately is a pool from which programming talent can be drawn.
OpenOffice has a large number of volunteers helping out in all sorts of ways, promoting Openoffice, creating artwork, etc. The non-developer community around OpenOffice is probably one of the largest around- but, and this is the point, the project is visibily suffering from a lack of developer manpower, progress in OpenOffice is like the moving of glaciers. Just think how much better it would be if they could draw from such a large and diverse community
like Linux.
Sun has undoubtedly made tremendous contributions to FOSS. I and many, many other trully appreciate this. But Sun has done virtually nothing to grow FOSS communities. One could even argue that this is not their job-and one would be right-it is not Sun’s job to create FOSS communities around it’s products. The issue here is whether Sun is actively precluding the development of such a FOSS community around OpenOffice by virtue of it’s propietary hold. Sun ought to Open the cage and set this bird free.
(lol, I can see some abiword/koffice devs laughing at setting OpenOffice free as in uncaging a bird-a bird ? that damned thing is so heavy that it couldn’t fly even if it had anti-gravity….
“If IBM was steward of this code and managed it in the same fashion Meeks would probably be lambasting them for their mismanagement.”
If IBM was steward of this code… look at Lotus Symphony… nobody would be lambasting it, because it would be proprietary.
“Novell has over 20 years involvement with Microsoft-whether as a company producing software to run under windows, or standing before a judge in legal issues with Microsoft.”
That’s not the current issue… the problem at hand involves SUSE, and GPL code… Sun also works with Microsoft in areas such as virtualization… but doesn’t hijacks GNU/Linux and exposes it’s users to the risks of legal litigation from MS.
“Expecting Novell to completely purge any relations to Microsoft after having purchase SuSE and Ximian is like expecting IBM to divest from the mainframe business once they started focusing more on software as a service product.”
That is not expected.
“Moreover I highly doubt that Meeks was consulted by the higher-ups at Novell when they made deals with Microsoft in the past couple of years. You seem to conflate the corporate political issues between Novell and Microsoft with the *agendas* of those employed by Novell who are working on Free Software.”
True, never stated the opposite of this. But maybe Meeks should talk to his higher-ups about their agendas before attacking other companies and their decisions on their products.
“Although I do not have access to any meaningful statistics about who has historically contributed the most to OpenOffice”
Well, of course, Sun.
Even if Sun didn’t produce StarOffice’s original code, they did pay for it, modified it and open sourced it… and keeps working and paying for it’s development… The fact that they didn’t originally made it is irrelevant.
“Sun maintains a propietary hold on this Free Software through a) all the hurdles it creates for contributors b) copyright assignment.”
The GNU project (most of what constitutes the Linux world right now… the Linux kernel being just a VERY small fraction) also requires copyright assignment and hasn’t stopped anyone from contributing. For better understanding of why this is necessary (and also applies in Sun’s case), read: http://www.gnu.org/licenses/why-assign.html
“Now why did Sun release the code for OpenOffice under a FOSS license? I don’t really know the answer to this”
Does it matter?
“community and corporations which build on FOSS(like IBM)”
IBM doesn’t build on FOSS by pure altruism, IBM sells whatever can be sold and can give services to, that is all that matters. Don’t expect an open source AIX, DB2 or Lotus soon…
“Mozilla is struggling with many of the same issues which OpenOffice has-a horrifically complex codebase which scares off many potential contributors, but a) this situation has gotten noticably better in the last years b) they also have spawned a vibrant community of plugin writers-which ultimately is a pool from which programming talent can be drawn.”
same thing happens with OpenOffice’s extensions: http://extensions.services.openoffice.org/
Edited 2008-10-12 03:51 UTC
I was just reading this and I thought it would be interesting to share… it also has some things Mr Meeks should consider when doing his analysis: http://blog.vaheder.com/?p=46
“2. Linux software is free, as all software should be
Linux software, for the most part (as far as I know), is quite free. There are packages like OpenOffice which prove that the open-source model for software development does, in fact, work. There are unique software creations, such as Beryl or Compiz-Fusion that do things that no Mac or PC has ever done with regards to 3D desktops. However, these are but the exceptions to the rule in Linux development.
OpenOffice is an anomaly.
First of all, it~Ac^AEUR^ATMs a software package created, for the most part, using the machinery available to a large software company; namely, Sun Microsystems. This means that all proper development tactics, such as planning, developing, debugging, and patching is done with corporate efficiency, something scoffed at by the larger Linux community. This lies in stark contrast with typical development seen in Linux, repeated again and again on Sourceforge. Software projects that barely got out of beta that have one or two maintainers at most that updated their software less and less frequently until it either ~Ac^AEUR^Aoedies,~Ac^AEUR^A or is integrated into another software package that, more likely than not, will die the same death in turn. This often happens with productivity software in general, and this is not a trend one often sees with packages that deal with services on Linux, which, as it turns out, is actually a very, very good thing.
Secondly, OpenOffice is Sun Microsystems~Ac^AEUR^ATM probono effort. Sun sells servers (and, by extension, Solaris), so more than likely OpenOffice is just a tax break for them. Sun makes no money (that I can see, as there~Ac^AEUR^ATMs no advertising or anything like that) since there doesn~Ac^AEUR^ATMt seem to be any revenue stream from OpenOffice. How many purely software companies (like Microsoft, for instance?) could possibly survive given this business model?”
Edited 2008-10-11 23:28 UTC
If someone doesnt like what SUN is doing, it is easy to fork! Try that with Microsoft or any other large company that does something you dont like.
No matter what, SUN opening up all their software is a good thing. I wish all other large dinosaurs would follow. Imagine MS, IBM, Oracle, SAP, etc following in SUN’s foot steps and opening up everything. If you didnt like something, you could just fork! Wouldnt that be a good thing? Do you rather cope with companies that has closed source, or companies with open source?