Contemplating the possible retirement of Apache OpenOffice

[Villeroy]

http://lwn.net/Articles/699047/

[Hagar Delest]

Indeed,

AOO is at serious risk of dying, unless we get more volunteer developers, especially C++ programmers.

OK, this is a bit harsh but you need to be aware that (to make things short) the Apache Software Foundation Board considers that the AOO project is not thriving enough (poor reactivity to fix security issues).
A message has been sent to the list to investigate the case AOO were to be sent to the ASF "attic", that is its development set at full stop. Of course this is not the case yet. The discussion will certainly create some FUD so better be prepared.

The discussion should not be taken as a prophecy but more as kind of cry for help. Either it creates an electroshock and the project manages to gather the energies required or it is the end.

The topic: [DISCUSS] What Would OpenOffice Retirement Involve? (long)
And to follow: What would OpenOffice NON-retirement involve?

The news is already starting to spread: OpenOffice, after years of neglect, could shut down.

So, if you know people or companies who could help, don't hesitate.
For the record: Get Involved
A new mailing list was created. You can subscribe to recruitment@openoffice.apache.org by sending e-mail to recruitment-subscribe@openoffice.apache.org

One thing I don't see is if the poor momentum of the project comes from the ASF way, from the PMC or from the Apache v2 license. Meaning: what needs to be changed to create enthusiasm again?

[Daiwe01]

I think the current situation lies wholly at the feet of the inept Project Management Committee (PMC).

I had hope when Iverson brought his enthusiasm and work ethic and became the PMC chair (but it was a close vote even then). His subsequent resignation from the project hinted, to me, at an increasingly dysfunctional PMC. Now Schenk, a 15-year OO developer, has resigned from the PMC and the project less than two weeks after volunteering to become the 4.2 release manager.

Hamilton (current PMC chair) seems focused on copious paperwork posts. Perhaps he, and other Apache Software Foundation members, should devote energy to Managing the Project and displaying positive efforts toward enhancing the development environment.

[orangeli]

That Hamilton dude is such an A$$h01e.
I despise apache software foundation.

[Hagar Delest]

If he hadn't done that, I doubt anyone would have taken the ASF board message seriously (who had heard of it exactly?).
There are times when a major crisis is necessary for a project to chose between death and revival. The moment has come. Of course it'll be bad because much FUD will be spread. But on the other hand with all the fuss that it will create, this may be the real and last opportunity for AOO to get momentum again.

[floris v]

That Hamilton dude is such an A$$h01e.
I despise apache software foundation.

How ironic. You are, I suppose, aware that most web servers run Apache software, so that you can post this here enabled by Apache software, right?

[pescetti]

As I wrote yesterday to the users list: this was an exceptionally bad e-mail from Dennis, that was interpreted in an exceptionally misleading way.

Dennis was playing a "what if" game seeing shutdown as an option (just one possible option) that a project with low activity must be prepared for, even it if never happens. So, in his usual style, he simply outlined all the steps that one should take in case.

It was immediately clear to me and others that this would have had two bad consequences:

1) Some press reading what they wanted to read and ignoring that this was a "fire drill" if you wish, but not an actual plan

2) Some active developers feeling insulted.

Despite the clear consequences, I'm sure that the communication from Dennis was well meant.

Another thread on how to make sure that the project is not retired was started and messages there now outnumber the other thread.

Code commits never stopped. My hope is that everyone will be back and that in a few days we can continue work towards the next release as usual.

[Hagar Delest]

I agree with you Andrea (pescetti) but this is a serious alert for the project, especially regarding its ability to deliver a secure software when needed (this is the root cause).
It also highlights again the gap in the attractiveness between AOO & LibO.

I also find that there is very poor communication from the project to the users. There is no clear direction. If there are still commits, to they relate to new features? Bug fixes?
Even if I'm subscribed to the dev mailing list, I don't see anyone following a common vision.
For example, it would be interesting to dedicate a release following a bug fix session with popular bugs and to tell about that. I mean something that gives users a reason to wait for a next release with expectations.

[orangeli]

How ironic. You are, I suppose, aware that most web servers run Apache software, so that you can post this here enabled by Apache software, right?

I don't care!
You can add "the programmers are from ASF" under Hell in:

[Hagar Delest]

Please, this is not a troll about what devs do.
The problem is not with the devs, it's more with the lack of them. I say lack of them because the security question triggers a warning so I guess there was not enough manpower to get the fix in due time.
however, the root cause may be elsewhere. ASF way? PMC management? Other?
The lack of communication/marketing let the outer spheres believe that the project was in limbo (idea well fueled by dedicated people in parallel also). We should learn from that "crisis" and give a signal to the users to say we are still there and we have a vision, which is... something to be agreed on, and the project commits to pursue this vision.

[pescetti]

its ability to deliver a secure software when needed (this is the root cause).

The link with security is because just a couple days earlier we had released a (not critical, hence we simply supply a fixed library) security update. But sure, it goes without saying that in all maintenance release we fix the outstanding security bugs.

I also find that there is very poor communication from the project to the users.

I couldn't agree more. I'm not exaggerating in saying that we need good communication people more than we need developers.

There is no clear direction. If there are still commits, to they relate to new features? Bug fixes?

Both, but since everyone is a volunteer we pick our priorities according to time, skills and personal preferences. In this respect, it is normal that you can't see a definite vision. That is normally part of the release planning (so: when a new release approaches, volunteers as a group choose priorities). It has become of course very hard to do something like that for the next release now.

[Hagar Delest]

Yes but for the last point, perhaps we could try to have a consensus to launch a bug fix session (for example) to show that devs can put up with a set (common) direction.

For the security issue, I just skimmed the messages on the dev list but I can't find any information given to users: are there instructions given? Are all platforms impacted?

[pescetti]

For the security issue, I just skimmed the messages on the dev list but I can't find any information given to users: are there instructions given? Are all platforms impacted?

Yes, (to dev list, to users list and to announce list) but everything is buried in monster threads and overlong messages with longer-than-needed explanations. The issue itself is https://www.openoffice.org/security/cve ... -1513.html and we do not consider it critical (no known exploits available; also, anti-virus software would block a malicious file trying to use this vulnerability). It was not disclosed with a new release -as it would be normal- since the company that discovered it would not wait for the next release, so we published the patch instead.

[acknak]

...
I also find that there is very poor communication from the project to the users. ...

This was the core concern when Oracle decided to hand off to ASF: ASF is a developer organization. They depend on other entities to take the software ASF develops and turn it into actual products and interact with the users. In the case of OO, developers are just one part of the needed effort. The user-facing or community-facing pieces have not been well managed because that's not what ASF does. It just isn't a good fit for a project like OO.

On top of that, actually contributing to the product has huge barriers; I wonder how many people find OO fun to work on. At least the LibO people seem to be having fun. Well, we have fun here, too, for the most part.

[Hagar Delest]

Thanks Andrea, that's clearer now. I've update the relevant topic that had been forgotten for a long time: Vulnerability bulletin.

On top of that, actually contributing to the product has huge barriers; I wonder how many people find OO fun to work on. At least the LibO people seem to be having fun. Well, we have fun here, too, for the most part.

+1. That's what I mean when I talk about the ASF way.
But not sure if it comes from the process or from the people.

It's clear that if adjustment is to be expected, contributors are not that ready to do so (they move to LibO). Knowing the specific scope of AOO as an end-user application, perhaps the project should negotiate those needed adjustments with the ASF. Just my 2 cents here but we need at least to analyze why the project doesn’t manage to attract contribution (that's at least what is our feeling).

And yes, glad we still have fun in the forum!

[Lupp]

As you easily can read from my signature, my interest in the topic is different.

I am neither well informed nor much interested in details concerning the disputes that led to the 2010 schism. I gravely regret it, and I claim a sheer necessity to revert to consensus and unity.

Despite the fact that I also have some preferences about the ways free software in the range of office applications should develop, my main concern is that it can develop at all over a long time, and finally (20+ years from now ?) can overcome the bribable variant(s) doing everything under the sole objective of commercial success.

Reliable long-term access to documents of some standard types and reliable - basically unchanged - functionality, including the already neglected access to "historical" (older than a very few years) documents is indispensable. Things cannot be left to the goodwill of money harvesters. And even now you can no longer expect an ods-document to produce the same results under any version of LibO or of AOO (even only upwards).

Independent development of more than one branches of free office software can not ensure the needed conformity and quality. It can not at all manage repairs concerning the damages done within the last few years. Least of all it can process the huge slag piles of bugs separately "cultivated" for each branch, not to talk of the reanimation of low ranging (?) components like 'Math'. QA must not be just a slogan.

Apache worry about a lack of developers, QA folks, and other specialists? So do TDF. Unite the efforts!

[kingfisher]

I have read through the second thread mentioned by Hagar Delest in the second post of this thread.

It seems that sufficient changes have been approved to warrant a new release of the software so it seems to me that the project is still alive. The thread highlights the need for more developers (a need apparently shared by LO) and the need for hardware.

I for one consider slow and infrequent releases as a positive attribute.

Talk of the project's demise is exaggerated. It is a good time for users to consider making donations or other contributions.

[Hagar Delest]

Donations are not easy IIRC because it would go to the ASF and they cannot redirect it to a dedicated project.
Code contributions would be the best of course.

The changes may be there now but what I understood is that it took too long to deliver them.

[orangeli]

No donations? another reason for OOo and ASF to split!

[kingfisher]

Donations are not easy IIRC because it would go to the ASF and they cannot redirect it to a dedicated project.
Code contributions would be the best of course.

The changes may be there now but what I understood is that it took too long to deliver them.

Is it worth posting some call to arms on the forum?

[Hagar Delest]

I think so. The discussion on the mailing list shows for example that there is not enough support for Mac builds.
Moreover, we can't ignore the fact that the ASF board was bothered enough to require some action.

In fact, I thought that there would be more talks about that in the forum from users fearing the end of AOO.
It may be much ado about nothing but it's a good opportunity to give visibility to the project, to confirm that it has never been dead at all and to have new devs on board (some already have proposed to help on the mailing list).

[jeehyun]

Donate name to Document foundation.
or sell OO to IT company like softmaker.

[domesticatedzebra]

Please, this is not a troll about what devs do.
The problem is not with the devs, it's more with the lack of them. I say lack of them because the security question triggers a warning so I guess there was not enough manpower to get the fix in due time.
however, the root cause may be elsewhere. ASF way? PMC management? Other?
The lack of communication/marketing let the outer spheres believe that the project was in limbo (idea well fueled by dedicated people in parallel also). We should learn from that "crisis" and give a signal to the users to say we are still there and we have a vision, which is... something to be agreed on, and the project commits to pursue this vision.

I realize that this thread has been inactive for a while now, but since you asked, I thought I should provide some input.
As someone who currently deals with ASF software as part of his job, I can write an entire novel describing my experiences.
However, of all the genres of literature in existence, that novel would most closely resemble a horror story.
In particular, it would be a Stephen King novel, what with all of the unexpected bugs and behavior I've encountered.

We can spend the rest of the month pointing fingers at the ASF leadership and the PMC overseeing this project, and it won't get us anywhere.
At the end of the day, it is the developers who are responsible for writing the code and delivering the product to the users.
If there is a shortage of developers, the solution is simple: give more people the ability to contribute code.
I'm not an expert on the OpenOffice development process, but from what I've seen, you have to be selected by existing developers to get commit privileges:

I'm not a committer, but would like to contribute, what should I do?

Apache OpenOffice welcomes your contributions!

To contribute to the OpenOffice product, you will need to submit subversion patches ("diff" files), and attach them to issues in our issues reporting system, Bugzilla.

Your contributions will be recognized by the committer submitting the PATCH in the log entry for the committed patch, and you will be added to the OpenOffice credits page to recognize your contribution.

The following sections will explain more about this.

Which leads to an interesting paradox. In order to contribute code, you need to be a member of a closed community of contributors.
If your project does not allow anyone with the necessary technical skills to contribute code, you are not open source. Period.

[floris v]

(...) Which leads to an interesting paradox. In order to contribute code, you need to be a member of a closed community of contributors.
If your project does not allow anyone with the necessary technical skills to contribute code, you are not open source. Period.

Open source doesn't mean that anyone can contribute code. Then it would be wiki source. It simply means that the source code is public and can be viewed by anyone. It certainly doesn't mean that anyone can make sense of it, by the way, but that's another story.

[domesticatedzebra]

(...) Which leads to an interesting paradox. In order to contribute code, you need to be a member of a closed community of contributors.
If your project does not allow anyone with the necessary technical skills to contribute code, you are not open source. Period.

Open source doesn't mean that anyone can contribute code. Then it would be wiki source. It simply means that the source code is public and can be viewed by anyone. It certainly doesn't mean that anyone can make sense of it, by the way, but that's another story.

Come on, you probably know Wikipedia is as accurate as Encyclopedia Britannica:

In order to test its reliability, Nature conducted a peer review of scientific entries on Wikipedia and the well-established Encyclopedia Britannica.

The reviewers were asked to check for errors, but were not told about the source of the information.

"Only eight serious errors, such as misinterpretations of important concepts, were detected in the pairs of articles reviewed, four from each encyclopedia," reported Nature.

But seriously, what about all the projects on GitHub? Are they "wiki source" because there are fewer barriers to contributing to them?

[floris v]

@domesticatedzebra: you completely missed my point. Which is: that "open" does not refer to who can contribute to the source code, but to who can access it.

Come on, you probably know Wikipedia is as accurate as Encyclopedia Britannica:

Your words, I didn't refer to the quality of Wikipedia, but to the statement that anyone can edit it.

[henke54]

The end of OpenOffice will be a good thing for open source and for users.

Let me explain.

[Lupp]

(Coming from StarOffice {StarDivision / SUN}. Gone to the 'LibO' branch early. Never abandoned AOO definitely. Talking about LibO here. )

As my signature is showing already for a long time now, the current way of things might comfort me?
Not quite.
I certainly am not an insider nor a technician, and being retired from my job I am not even any longer one who is using LibO "in production". Just an experienced user of some of the modules spending some time on free software contributing to forums and marginally taking part in the fight against bugs (a few suggestions about how to fix one included).
I do not know anything about 'Impress' (virtually never used any brand of that kind of software) and am not too much interested in 'Writer'. My main fields of former professional usage were 'Calc', 'Draw' and 'Math', and these are still in my focus.
Most of my activities contributing to https://bugs.documentfoundation.org are concerning Calc. Thus I have a probably valid impression of the proceedings there. In short: It's a mess generally - and surprisingly fast in a few cases when a developer decides to choose the "strategy" of cherry picking. At the same time there doesn't pass a day that I not experience new bugs, in some cases so confusing that the mere preparation of a description and a demonstration for a report is beyond my capablities.
From my observations regarding 'Calc' I suspect development to no longer stick to well proven rules of software engineering. With respect to 'Math' I suspect there is no development at all. (See https://ask.libreoffice.org/en/question ... e-bracket/) Concerning 'Draw' there seems to be a campaign considered. (See https://design.blog.documentfoundation. ... from-draw/ ; I am very sceptic about it.)
The UI? The 'Help' system (Offline / Online )? The "official" askbot forum (Not a forum at all!)? Important fields of profound criticism!

Thus:
Let's earn improvements of the situation! Let arise the powerful UFO a United Free Office would be.
Don't go the way of LibO without any changes! More thorough considerations needed; less versions and less confusion, too!
Actually merge with what still is functional in AOO development, human resources and concepts as well. There surely are nuggets!

[witstert]

Surely, AOO should take the route that Microsoft is, apparently, taking, according to the latest "Tweakers" e-zine, that "Microsoft has released the Office app for Windows 10 in its Microsoft Store." and make AOO or an online version available through Google Play Store and the Apple Store?

Oops! Left off the URL for the "Tweakers" article: https://tweakers.net/nieuws/149438/micr ... -21&u=1300

[kingfisher]

The techrepublic article is dated September 14, 2016. I misread that date, thinking it was a recent article. I am not familiar with the language used in the tweakers article.