Forward message because the mailing list was dropped
-------- Original Message -------- Subject: Re: Slashdot Article Date: Thu, 29 Aug 2013 10:13:27 +0200 From: Jürgen Schmidt <jogischm...@gmail.com> To: Keith Curtis <keit...@gmail.com>, keit...@gmail.com On 8/29/13 1:37 AM, Keith Curtis wrote: > Hi Jürgen, > > Thanks for your thoughtful mail. I am glad to know you want to re-unify > the projects. The more others see things like you, the faster it will > happen. > > I believe it matters how much code comes from IBM. Code coming from > anywhere is good, but diversity of community is a good measure of > health. LibreOffice, the Linux kernel, and other groups publish such > charts for a reason. In AOO, it is hard to find out where everyone > works, which can confuse many into seeing more diversity than what exists. > > Also, since you work for IBM today, you could have a bias. I personally > don't entirely trust IBM to have the best interests of free software: > because of AOO and other products. I was amazed to learn 5 years ago > that less than 10% of IBM employees ran Linux. I don't know what the > number is now, but I doubt it is above 25%. They are a massive company, > but even assuming that they had good intentions and execution, it would > still be a risk. You might not find the current situation a problem, but > at some point, perhaps around 95%, you would recognize it. It is good > that IBM ended the Symphony fork. You might ask yourself why this was > done, I think the success of LibreOffice was the primary impetus. I don't know for sure it was before I joined IBM but I don't believe that LO played any role here. And this point LO was probably known only by the people who have created the project. > > The Apache license is a popular free one, but I'm of the opinion that it > is inferior in general, and inferior for this codebase. The former is a > topic out of scope here, but here is an article talking about the > latter: http://ebb.org/bkuhn/blog/2011/06/01/open-office.html Examples > of successful projects built under an Apache license are not concrete > proof. Linus Torvalds has said that making Linux GPL was "the best thing > he ever did." Copyleft protects the small contributors. We aren't going > to resolve our disagreement here, but the choice of the Apache license > has made the situation with LibreOffice more difficult to resolve. So I > think AOO should consider whether the lax license is so good that it is > worth the massive downsides. The LO code is probably >90% under the Apache license because you can't remove the license. You can only put your significant changes under a different license. Removing German comments is of course not significant. Anyway a simple fact that is not mentioned by some people for obvious reasons. So please do your homework and figure out some facts on your own and don't listen to other people only. > > It is true that the 65M downloads is a reason to work in AOO. However, > there are many other reasons to choose one or another: > http://mmohrhard.wordpress.com/2013/01/30/why-i-contribute-my-changes-to-libreoffice-and-wont-relicense-them-to-a-non-copyleft-license/ > Contributors don't work with random users, they work with other > contributors. In various ways LibreOffice has a superior codebase. For > example, most of the German comments have been removed. For you this > wouldn't be a big deal, but for others you can imagine that working in a > codebase they can understand is more important than working in a > codebase that is more popular. That is just one example, there are many > other factors to consider other than just number of users. Another is > the fact that LibreOffice has better tools. Others can explain this in > more detail more than I, but if you haven't been following all their > infrastructure work, you'd be blown away by it. yes, great work removing or translating German comments that are not useful at all. It's a cosmetic change only and is of course nice but no real value for any developer who can read the code. Many comments in the old OpenOffice code are really useless (which doesn't make it better). > > LibreOffice 4.1 also has hundreds of user-visible features that AOO 4.0 > does not. There is the most popular codebase today, and there is the > most popular one tomorrow. Also, some might believe that the fork will > eventually end. So why not just associate with the codebase you think is > superior while you wait for that to happen? Most contributors don't care > about the name as much as the community. It's quite simple I don't like the setup and the structure in the project. I saw people that I know for many years moving into positions at TDF and now control the project. In the same way they didn't liked before when Sun/Oracle was the steering instance. I was offered a position in the steering committee when I would join LO and everything should be forgiven. I was really surprised what should have been forgiven because I am a strong supporter of OpenOffice since the beginning and part of the project much longer than most of the LO project members. I was of course not interested in any position and I am still not interested in one. I am happy here at Apache where all are equal. Well I am acting as the release manager but if somebody else is interested to do it, it's fine for me. I don't have respect for the way how LO is taking our code. It might be ok from a legal perspective (still not 100% proven) but from a ethically pov it is not and I don't support it. And it is the double moral that annoys me to claim public they do a rebase once to be able to relicense (which is true otherwise relicensing wouldn't have been possible at all) and now they continuously cherry pick fixes and features from a project that is dead and don't move forward. Mmh doesn't really fit together. > > It is true that it will get harder over time to port changes from AOO to > LO. However, at least it is possible, whereas the reverse is not. This > makes the situation worse than for example Ubuntu / Debian where code > changes can flow both ways. Furthermore, it is much cheaper to port a > feature over than to design and write it. I estimate it is at least 20 > times easier. There will be bugfixes port over as well, but LO makes > frequent releases, so they will also get absorbed soon enough. It is > much easier to port a bug fix than a feature. > > In addition, the Sidebar is just one feature. It will be stable and > better in LO 4.2 in less than 6 months. Most of AOO 4 features came from > Symphony. Once they are absorbed in LO, the question of the cost in > porting things really has to do with the ongoing rate of change in AOO > over time. Given it is much smaller than LO, and it is much easier to > port than to write, I don't think it will be a problem. > > I'm glad to hear you want to continue the power and success of > OpenOffice. I hope you are concerned which codebase would be the better > one to represent that brand. AOO can of course benefit from some changes/fixes in LO and any individual developer is invited to make his single contribution available under the ALv2 as well. In the long term I believe we will have the better code base. And yes LO has at the moment the cleaner build system (completed the switch to gmake) and they have some good tools in place but all this is possible at Apache as well if their are people who want to drive such things. Juergen > > -Keith > > > On Wed, Aug 28, 2013 at 3:40 AM, Jürgen Schmidt <jogischm...@gmail.com > <mailto:jogischm...@gmail.com>> wrote: > > Hi Keith, > > I put you in cc because you are not subscribed > > On 8/28/13 6:45 AM, Keith Curtis wrote: > > One of the points I would like some clarification on is in the article > > defending ASF by Andrew C. Oliver: > http://www.infoworld.com/print/225555 > > > > In it, he wrote: "Nearly all of the very active Apache OpenOffice > > developers work for IBM directly or indirectly." > > > > Can someone explain more about this? For example, is there a table > showing > > who everyone works for? From that, one could make a chart showing > how many > > of the code changes in AOO were made by various companies. LibreOffice > > publishes such charts: > > > http://documentfoundation.files.wordpress.com/2013/07/developers6.jpg From > > my research, I'd guess AOO 4 was at least 80% IBM, but I've not > been able > > to find out information on several people so it is an estimate. > > > > In addition to regularly posting information about the number of user > > downloads, it would be great to regularly post a chart that shows the > > diversity of the code contributors. Everyone who is a volunteer is > allowed > > to choose which community they'd like to join and this would give them > > important information. I know there are IBM employees all over the > place > > working on release management, QA, and other areas, but just knowing > > diversity about code changes would be useful. > > > > If it were 90+% IBM, they could end the fork very quickly. The > remaining > > 10% would probably move, but even if they didn't, it would still > be much > > better. Ending this fork is one of the best things that could > happen to > > free software today. As a former Microsoft Office programmer, I > can state > > that this fork benefits them immensely, helping both Office and > Windows. I > > realize that no one here wants to help Microsoft, but it is an > "unintended > > consequence." If I still rooted for Microsoft, I would be laughing > at the > > incompetence because it indirectly gives them billions of dollars. > As I > > root for Linux now, I find it sad because it is a lot more work to > build > > two communities and brands: http://keithcu.com/wordpress/?p=3163 > > > > thanks for sharing your view and insights. And I totally agree > reunifying both projects would be the best to ensure a powerful and free > office productivity suite in the future and focus together on the > challenges of the future which is not really the desktop. > > I believe it doesn't really matter how much code contributions come from > IBM. More important is the fact that IBM stops the Symphony fork and > joined Apache OpenOffice to work together with the community (and > indirect with other companies) on one common goal and one open source > office productivity suite. And contributed the source of Symphony as > well to Apache OpenOffice. > > Oracle decided where the project should continue and granted the code, > the copyright, the brand etc. to Apache. The Apache license is a good > one and has proven that it works well for open source projects. Android > as a very prominent example! > > Apache OpenOffice will continue and we are open for everybody who is > interested to join the project and help to bring it forward. A really > flat structure where all project members are more or less equal. > Decision are made on the public developer list and not in the PMC for > example. > > > 65 million downloads in 1 year and that are only the officially > counted downloads. Many more on other not counted download servers. > > Any individual developer should decide in which code base a bugfix or > feature should go to benefit the majority of office users. > > LibreOffice is taking our code which is fine but we have still more > users and the cherry picking of fixes or other improvements will become > more difficult. Take for example the sidebar, I don't want to know how > much time they have spent to integrate it and it is less stable. Anyway > we know the underlying framework because we have implement it from > scratch and the sidebar was not only a simple merge as some people try > to explain where the code comes from. All the migrated panels were > reworked to fit in our new framework. > > I hope you have send the same or a similar message to the TDF or the LO > project ;-) > > We are here to continue the power and success of OpenOffice and again > everybody is invited to join. We can't promise any position because we > don't have them and we have only a flat hierarchy with one project > chair. But the chair has not more control in the project as any other > contributor. > > But it's a good pace for anybody who simply want to drive the project > and want help to bring it forward. > > Juergen > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
