I've felt for a long time (maybe even a decade) that Apache is the place OSS goes to retire/die. It is heartbreaking in some ways, they became the stewards for a lot of interesting projects [1]. And the venerable web server was so important to the early web. I am still running Apache vhosts on a few personal websites.
It is also a totally unfair characterization. I mean, they still house relevant projects including Kafka, Lucene, Zookeeper, Spark, Arrow and many others.
But it does feel Apache lives in a kind of stasis, a relic of the "old" web, like geocities sites or MySpace. If a project ends up in Apache it feels to me like it has gone out to pasture.
It seems that “younger generation” programmers underestimate the value of governance in OSS. The recent Terraform kerluffle couldn’t have happened if Terraform was in the ASF, etc.
Agreed... We are seeing dozens of database companies turn their OSS into basically proprietary shared code licenses. And we seem to still pretend its "open source"
Collaboration, across companies, not ownership by one company is the root of open source. To do that you need some kind of governance model. Essentially "ownership" by one company is conducive to quick progress on specific problems, but not collaboration.
You are assuming that a strict and stable governance model is good for OSS.
But often it is the least stable projects with lots of hostile forks and huge mailing list arguments which turn out to be the most successful projects.
Strict and stable governance model is perhaps dull and turns away contributors who feel they will never get to the top of such a stable project with long timelines and complex procedures for everything.
There is a quote: "When you stop growing you start dying" (William S. Burroughs). I am not sure I believe that the kind of governance that Apache is providing is helping to grow the projects under its supervision.
Your comparison to Terraform is also interesting. Apache Kafka, for example, has been absorbed by Amazon (MSK). Apache Spark is Amazon EMR. Just looking through Amazon's paid service offering I find several Apache projects rebranded and for sale. A cynical take could be that Apache is an org that helps companies like Amazon profit off of the work of open source developers.
Apache has lots of active projects. But they are mostly data analysis. At work, we use Airflow to run Spark jobs which uses Hadoop and Hive.
Apache also has a lot of dead projects that never got picked up by users. Apache should have a marker for zombie projects so people don't start using them.
Software projects have a natural lifecycle, and since we've been around for almost 30 years, there's going to be a number of projects in that state.
However, the ASF doesn't force projects into the attic while they still have a community around them. OpenOffice is an edge case - there's still a project community, primarily around templates and other end-user-centric activities. They're just not doing much with the actual software any more.
Granted, there's room to disagree as to whether 1) the ASF should force projects into the attic and 2) whether OO actually has an active community. But the Foundation's position is that this is the decision of the project, not one to be forced top-down by the Board.
They do have the Apache Attic for that purpose, but they typically defer to the project maintainers to make the decision to retire a project, unless there are no longer any maintainers.
> If a project ends up in Apache it feels to me like it has gone out to pasture.
There are many Apache Foundation projects that are active and important or complete and important.
Unfortunately OpenOffice is not any of the above, and it was that way almost from the beginning. The original OO community largely went with LibreOffice. They did not like the interference of Oracle and later IBM in their project's governance. So, Apache got the trademarks from Oracle and a community of people who were largely not maintainers. This put Apache, almost from day one, in position to be competing with the community of one of their own projects.
> I've felt for a long time (maybe even a decade) that Apache is the place OSS goes to retire/die.
It's a bit too strong as others have pointed out that there are very active projects under the Apache umbrella. But I do feel that there's a core of truth in this. Specifically, Apache does appear to provide a "way out" for large enterprise customers that want to dump non-strategic (to them) code somewhere and not necessarily invest in it. Seems something that could be solvable on the ASF side by having additional criteria before accepting projects, and maybe require a certain level of activity for continued membership.
>> > Apache is the place OSS goes to retire/die
>> Fortunately many of us in the world disagree with the dystopian warped view.
They would still be doing everyone a favor if they shut down OpenOffice completely and hand the trademark over to The Document Foundation. The name is better than LibreOffice, and the fact that a grossly outdated version of OpenOffice is what people will be finding is detrimental to the project. Most the developers and the code have moved on, but the name and a bunch of old code lives on with ASF for no good reason. Seriously there is NO reason for it except the completely selfish (and harmful to others) excuse that ASF can claim to be the stewards of OpenOffice. In my dystopian imaginary world they are even paid to maintain this situation to prevent more widespread adoption of LibreOffice.
At work I've lovingly but sadly referred to Apache as the "Crazy Cat Lady of Software." From the outside it looks like they're willing to take on any project that someone wants to give away regardless of whether they have the capacity to provide maintenance / governance of it themselves.
While I do love the preservation aspect of it, seeing that a project is maintained by Apache rings the same "Sure, but is the project already dead" bells in my head that I hear when I see a project from Google. In both cases, the name brings the concern that a project might be on its way out.
I was recently searching around for a good enterprise search engine and wanted something OSS. I found Apache Lucene and so far I really like it. It's got documentation, a clear release schedule, and I know lots of other people use it. I was happy to hear that it's Apache because it probably means it won't die because one maintainer goes away. Maybe it's silly, but if I find a project is Apache it gives me a good feeling.
Of course some of their projects die. That's OSS. But that's not something you can solve for. Even paid software goes away after awhile.
Unless it’s one of their many durable Java projects, or the web server that bares its name.
I think some things were (are?) bad fits for the ASF and languished as a result.
If you do a lot of development in the JVM ecosystem though you’ll find there’s a lot of ASF projects that are actively used and maintained
Go to die, or go to provide incremental improvements? Maybe a bit of both? Either said, it's a good question.
Some projects are going to die, and I don't think that's the fault of the organization behind it. I'd love to see a more detailed analysis of all the current projects at Apache, and their growth/decline, and a breakdown of _why_ they're growing or declining. Maybe Apache is the problem, maybe it isn't.
A big part of the problem is that the Apache foundation has become sort of a dumping ground for OSS projects where the corporate sponsor doesn’t want to be responsible anymore. They pull the full time developers and donate the project. In this context it’s not surprising that these projects “die”, because they’ve lost most or all of their contributors.
I believe that the death blow which lead to the fork LibreOffice was the lack of trust in Oracle commitment to Open Office and open source in general, after Oracle acquired Sun Microsystems.
Sun indeed had largely reduced its involvement in the project in the last years, and there was the "Symphony incident" [1]. The death blow was when Oracle arrived, started the trial with Google over Java/Android, inherited MySQL and almost killed it, inherited OpenOffice and the future did not look bright anymore [2, in German]
At about the same Sun, Oracle, IBM et al were cooling on the idea of developing comprehensive office software suites (~2008–2012), online suites like Google Workspace and Zoho were arriving in force, as were supporting do-business-online-not-on-the-desktop tools like Wrike.
If the OpenOffice kerfuffles weren't sapping enough energy out of the ecosystem, that users could suddenly use free, no-install-ever, not-equivalent-but-still-fine online tools did the rest.
Used to be an enthusiastic OpenOffice and LibreOffice user and developer—not of the core tools, but as part of bigger document processing and publishing workflows. Even presented at the OpenOffice Conference in 2007 (Moit de gust, Barcelona!) But after Google Docs, Sheets, etc. ... have not needed OpenOffice or LibreOffice nor used them in anger for over a decade.
> I believe that the death blow which lead to the fork LibreOffice was the lack of trust in Oracle commitment to Open Office and open source in general, after Oracle acquired Sun Microsystems.
I mean... can you blame them? I would have forked too, if anything out of caution. The lawnmower has no feeling.
Well, my understanding is that right after Oracle acquired Sun it found itself with a free database engine it did not need because it would compete with its own "free" Oracle XE product offering.
There was a lot of uncertainty about the future of MySQL... The original authors forked MariaDB and MySQL for many months did not receive any update.
Apache's stubborn refusal to direct OpenOffice users towards LibreOffice has done immeasurable harm to the Apache brand and it's basically ruined the OpenOffice brand. A great example of the damage that can be done by poor stewardship.
Has it? I have used both libreoffice and openoffice, I had no idea openoffice was related to asf until I read your post. I have heard other talk about how people should use libreoffice, but it isn't apparent why.
Back in the day there was real interest in "OpenOffice" among non-geeks. Governments were mandating that documents be published in OpenDocument format. Microsoft sought to muddy the waters by publishing the confusing "Office Open XML" format (which they then never implemented).
And then OpenOffice the project just... stopped. No new features, and eventually no new releases. And by squatting on the brand, they prevented that momentum from transferring to LibreOffice.
>> I have heard other talk about how people should use libreoffice, but it isn't apparent why.
Because there is almost a decade of enhancements and improvements to LibreOffice that are not present in OpenOffice. It's simply a better tool suite and there is no disputing that. If you try you'd be trolling.
I think the continued co-opting of the name of a first nation as their "brand" despite repeated calls from stake holders to do otherwise has caused plenty of harm to the ASF brand.
It's no worse than Amazon being named after the rainforest and river despite not originating there, or Apple being named after the fruit despite not being a greengrocer. Names get reused, it's not a big deal.
Last time I checked it was a small organisation run by 2 or 3 people with no real standing to speak for anyone but themselves. And AFAIK that was also the only time a major objection was raised.
And look, if you dislike the name then by all means voice your objections and let's have the discussion. I dislike the reactionary "STFU" response, which just shuts every discussion down. What I also dislike is this kind of pretence, which also shuts down the discussion.
> Most of the commits here seams to be mainly white space changes to random files.
It might not be "learning" but rather an attempt to game GitHub (e.g. rack up a lot of commits to important-sounding repos). My understanding is go around using automated tools to submit PRs that fix whitespace and typos in order to do that.
The trunk branch, the Bugzilla list, the OS/2 support, the user forums, the multi-hour build process, the XHTML 1.0 compliant badge on the https://www.openoffice.org/ homepage .....
Many years ago I became proficient with the Linux kernel build process, and later regularly built Mozilla with my preferred configuration and so forth. Both processes take some doing. I thought building OpenOffice couldn't be much of a challenge, it wasn't an operating system after all. I can't recall if I ever got it done. If I did, it was certainly the most vanilla possible build, taking all the defaults for a bog-standard build.
Remembering when I spent three days compiling OOo from FreeBSD ports, to get a build that didn't work, then discovering that this was a known broken port. I started using the Linux binary in emulation instead.
Well, Linux is just kernel, and as such doesn't take so much time to build. A whole operating system (depending on what's in scope) would take hours or even days
> 3. Honestly, just who cares? So what if OpenOffice is just being minimally kept alive?
The post is about the fact that it is _deceptively_ minimally kept alive.
A significant part of the decision process involving OSS choice is measuring active maintenance. Surely, these commits exist to give a false impression here, which down the road can harm those using unmaintained software.
How is it deceptive? Are they bumping version numbers? Are they packaging new releases?
Edit: before I'm accused of not reading the article, I did, but I did not look at the actual OpenOffice download page. The current release, dated April 2023, includes the following, which seems substantially more than "whitespace cleanup" as TFA author claims, but seems quite in line with software that is being maintained but not actively developed:
Other Improvements/Enhancements
Writer: Dialog "Frame" in Writer / automatic size / changing shortcut keys (German)
Calc: Protect Table – Dialog too narrow for German heading string
Base: Dialog text in "Datenbank austauschen" (exchange database) dialog is cut (German)
Calc: Use the existing icon for "Remove Filter" in toolbar and menu
Calc: Text in cell comments is more readable
Calc: Added support for Excel 2010s DateTime type cells.
Bug Fixes
Calc: Rich text cells imported from XLSX files have their contents duplicated
Several fixes for importing OOXML documents
All applications: Import fails when the Relationship "Target" attribute in _rels/.rels has superfluous slashes
Calc: XLSX with omitted cell references opens with empty cells
Calc: no text imported from XSLX when xl/sharedStrings.xml has wrong case
Several fixes for importing MS Excel 2003 SpreadsheetML files
Calc: ss:DateTime cell value is loaded as either only a date or a time
Calc: fractional seconds are silently ignored during import
Calc: cell with ss:MergeAcross="0" gets an extra empty cell to the right
Calc: import corrupts references to columns that are multiples of 26 (eg. Z)
Writer: Importing apostrophes from HTML fails
Calc: Cut-and-paste between spreadsheets causes incorrect cell reference changes
Writer: Awkward Chinese (ZH-TW) numbering suffix when importing RTF document
Calc: Last CSV line is silently lost on import if last field is quoted and end-of-file is reached before closing quote
It damages both the Apache and OpenOffice brands. People who have heard of MS Office alternatives have probably OpenOffice in their mind and will attempt to use it. They will probably be dissatisfied and have a bad impression of OpenOffice, Apache, and OpenSource in general.
> 3. Honestly, just who cares? So what if OpenOffice is just being minimally kept alive?
The pretense that this project is alive and not a zombie puts millions of users in danger who are downloading insecure trash under the impression that this is a live project that's the OpenOffice they've heard so much about.
They definitely are relevant despite what the armchair experts say on HN. Do you use AWS or GCP or Fastly? These companies don't provide huge sponsorship dollars to Apache out of the goodness of their hearts. Many of the cloud services you use are backed by one or more Apache projects.
Okay, so we've known this is happening for a while. But why though?
First, I don't understand what's the ASF's position here. Do they have some political reason to keep kicking this can down the road? Is there money on the line? Is it just somebody refusing to give up?
And given the licensing allows, why don't they rebrand LibreOffice, rather than keeping a zombie of a project?
Or why don't they just make some sort of deal with LibreOffice?
I believe a key issue is that the license of LibreOffice is copyleft (I believe some mixture of LGPL and MPL) and so the Apache Foundation--which specifically focuses on code that is not copyleft (a focus I personally disagree with, but is at least consistent)--feels something is lost to the world if they truly give up on the one permissively-licensed open source office suite.
That's a good point I haven't seen brought up before when this topic comes up. That really does make a lot of sense explaining why Open Office still exists, but not why Apache is not being transparent about the situation.
The pointless commits combined with the lack of transparency feel a bit nefarious. I wonder if it's more about personal drama between the projects than anything else?
Then again, the Apache2-licensed OpenOffice is not lost to the world even if it sits in attic.apache.org. If and when someone wants to maintain it, it can be resurrected.
You're completely right! The issue is, this mandate is being made by corporate donors and influencers which is why ASF should be a 501(c)(6) not a 501(c)(3) that it is. It's probably breaking the law, but at least unethical.
>> I believe a key issue is that the license of LibreOffice is copyleft (I believe some mixture of LGPL and MPL) and so the Apache Foundation--which specifically focuses on code that is not copyleft (a focus I personally disagree with, but is at least consistent)--feels something is lost to the world if they truly give up on the one permissively-licensed open source office suite.
And here I thought they were just against OpenOffice, this interpretation pits them against Free Software more generally. I'm gonna run with that for a while and see how it fits.
The ASF does not tell projects what to do. It provides a process and a place, and gives projects autonomy. The project, not the foundation, decides to keep running the project. So to characterize the ASF as "kicking this can down the road" for some kind of inscrutable motivation is just not how things work.
The members of the OO project want to keep running their project. They have a group of people who keep showing up to do user support, documentation, and produce a huge community around their templates. They just haven't pushed any new features for a while. But they still are there, keeping the lights on.
The ASF is not a top-down organization. It's driven entirely by those projects. And if you look at the dev list - https://lists.apache.org/list.html?dev@openoffice.apache.org - you'll note that this issue gets raised on a VERY regular basis, and the project participants always say "no thanks."
As to "making a deal with LibreOffice", I would encourage you to read the mailing list, where that, too, gets raised with great regularity. There, too, the community sentiment is "no thanks." On both sides.
Readit News
It is also a totally unfair characterization. I mean, they still house relevant projects including Kafka, Lucene, Zookeeper, Spark, Arrow and many others.
But it does feel Apache lives in a kind of stasis, a relic of the "old" web, like geocities sites or MySpace. If a project ends up in Apache it feels to me like it has gone out to pasture.
1. https://projects.apache.org/projects.html
Collaboration, across companies, not ownership by one company is the root of open source. To do that you need some kind of governance model. Essentially "ownership" by one company is conducive to quick progress on specific problems, but not collaboration.
You are assuming that a strict and stable governance model is good for OSS.
But often it is the least stable projects with lots of hostile forks and huge mailing list arguments which turn out to be the most successful projects.
Strict and stable governance model is perhaps dull and turns away contributors who feel they will never get to the top of such a stable project with long timelines and complex procedures for everything.
It was cool to be a UNIX/FOSS zealot during the late 90's while still getting monetary support.
Writing M$ on the email signature and such.
Eventually I had to find a way to pay my bills.
Your comparison to Terraform is also interesting. Apache Kafka, for example, has been absorbed by Amazon (MSK). Apache Spark is Amazon EMR. Just looking through Amazon's paid service offering I find several Apache projects rebranded and for sale. A cynical take could be that Apache is an org that helps companies like Amazon profit off of the work of open source developers.
Deleted Comment
There are still a lot of projects at Apache.
Check their last annual report [1], especially the sponsors and the statistics sections.
[1] https://apache.org/foundation/docs/FY2023AnnualReport.pdf
Apache also has a lot of dead projects that never got picked up by users. Apache should have a marker for zombie projects so people don't start using them.
Software projects have a natural lifecycle, and since we've been around for almost 30 years, there's going to be a number of projects in that state.
However, the ASF doesn't force projects into the attic while they still have a community around them. OpenOffice is an edge case - there's still a project community, primarily around templates and other end-user-centric activities. They're just not doing much with the actual software any more.
Granted, there's room to disagree as to whether 1) the ASF should force projects into the attic and 2) whether OO actually has an active community. But the Foundation's position is that this is the decision of the project, not one to be forced top-down by the Board.
When I find something interesting on github, before deciding to use it, I check the number of committers and the commit history/frequency.
Anyone can do the same for Apache projects.
There are many Apache Foundation projects that are active and important or complete and important.
Unfortunately OpenOffice is not any of the above, and it was that way almost from the beginning. The original OO community largely went with LibreOffice. They did not like the interference of Oracle and later IBM in their project's governance. So, Apache got the trademarks from Oracle and a community of people who were largely not maintainers. This put Apache, almost from day one, in position to be competing with the community of one of their own projects.
It's a bit too strong as others have pointed out that there are very active projects under the Apache umbrella. But I do feel that there's a core of truth in this. Specifically, Apache does appear to provide a "way out" for large enterprise customers that want to dump non-strategic (to them) code somewhere and not necessarily invest in it. Seems something that could be solvable on the ASF side by having additional criteria before accepting projects, and maybe require a certain level of activity for continued membership.
Fortunately many of us in the world disagree with the dystopian warped view.
They would still be doing everyone a favor if they shut down OpenOffice completely and hand the trademark over to The Document Foundation. The name is better than LibreOffice, and the fact that a grossly outdated version of OpenOffice is what people will be finding is detrimental to the project. Most the developers and the code have moved on, but the name and a bunch of old code lives on with ASF for no good reason. Seriously there is NO reason for it except the completely selfish (and harmful to others) excuse that ASF can claim to be the stewards of OpenOffice. In my dystopian imaginary world they are even paid to maintain this situation to prevent more widespread adoption of LibreOffice.
While I do love the preservation aspect of it, seeing that a project is maintained by Apache rings the same "Sure, but is the project already dead" bells in my head that I hear when I see a project from Google. In both cases, the name brings the concern that a project might be on its way out.
Of course some of their projects die. That's OSS. But that's not something you can solve for. Even paid software goes away after awhile.
If you do a lot of development in the JVM ecosystem though you’ll find there’s a lot of ASF projects that are actively used and maintained
Some projects are going to die, and I don't think that's the fault of the organization behind it. I'd love to see a more detailed analysis of all the current projects at Apache, and their growth/decline, and a breakdown of _why_ they're growing or declining. Maybe Apache is the problem, maybe it isn't.
Sometimes it is the place projects go when their original author wants to kill them but doesn't want to say they are doing that.
Deleted Comment
I believe that the death blow which lead to the fork LibreOffice was the lack of trust in Oracle commitment to Open Office and open source in general, after Oracle acquired Sun Microsystems.
Sun indeed had largely reduced its involvement in the project in the last years, and there was the "Symphony incident" [1]. The death blow was when Oracle arrived, started the trial with Google over Java/Android, inherited MySQL and almost killed it, inherited OpenOffice and the future did not look bright anymore [2, in German]
[1] https://web.archive.org/web/20130927072255/https://lwn.net/A...
[2] https://web.archive.org/web/20131024094250/http://www.heise....
If the OpenOffice kerfuffles weren't sapping enough energy out of the ecosystem, that users could suddenly use free, no-install-ever, not-equivalent-but-still-fine online tools did the rest.
Used to be an enthusiastic OpenOffice and LibreOffice user and developer—not of the core tools, but as part of bigger document processing and publishing workflows. Even presented at the OpenOffice Conference in 2007 (Moit de gust, Barcelona!) But after Google Docs, Sheets, etc. ... have not needed OpenOffice or LibreOffice nor used them in anger for over a decade.
I mean... can you blame them? I would have forked too, if anything out of caution. The lawnmower has no feeling.
There was a lot of uncertainty about the future of MySQL... The original authors forked MariaDB and MySQL for many months did not receive any update.
And then OpenOffice the project just... stopped. No new features, and eventually no new releases. And by squatting on the brand, they prevented that momentum from transferring to LibreOffice.
Because there is almost a decade of enhancements and improvements to LibreOffice that are not present in OpenOffice. It's simply a better tool suite and there is no disputing that. If you try you'd be trolling.
Last time I checked it was a small organisation run by 2 or 3 people with no real standing to speak for anyone but themselves. And AFAIK that was also the only time a major objection was raised.
And look, if you dislike the name then by all means voice your objections and let's have the discussion. I dislike the reactionary "STFU" response, which just shuts every discussion down. What I also dislike is this kind of pretence, which also shuts down the discussion.
https://github.com/apache/openoffice/commit/d2a7b3cc90e95392...
> https://github.com/apache/openoffice/commit/d2a7b3cc90e95392...
From that link:
> Most of the commits here seams to be mainly white space changes to random files.
It might not be "learning" but rather an attempt to game GitHub (e.g. rack up a lot of commits to important-sounding repos). My understanding is go around using automated tools to submit PRs that fix whitespace and typos in order to do that.
The trunk branch, the Bugzilla list, the OS/2 support, the user forums, the multi-hour build process, the XHTML 1.0 compliant badge on the https://www.openoffice.org/ homepage .....
Reminds me of the good old days.
Many years ago I became proficient with the Linux kernel build process, and later regularly built Mozilla with my preferred configuration and so forth. Both processes take some doing. I thought building OpenOffice couldn't be much of a challenge, it wasn't an operating system after all. I can't recall if I ever got it done. If I did, it was certainly the most vanilla possible build, taking all the defaults for a bog-standard build.
2. What has the author done to engage so far? Has Apache responded to their emails? Have they opened any threads in chat forums?
3. Honestly, just who cares? So what if OpenOffice is just being minimally kept alive?
The post is about the fact that it is _deceptively_ minimally kept alive.
A significant part of the decision process involving OSS choice is measuring active maintenance. Surely, these commits exist to give a false impression here, which down the road can harm those using unmaintained software.
Edit: before I'm accused of not reading the article, I did, but I did not look at the actual OpenOffice download page. The current release, dated April 2023, includes the following, which seems substantially more than "whitespace cleanup" as TFA author claims, but seems quite in line with software that is being maintained but not actively developed:
Other Improvements/Enhancements
Bug FixesThe pretense that this project is alive and not a zombie puts millions of users in danger who are downloading insecure trash under the impression that this is a live project that's the OpenOffice they've heard so much about.
They haven't been relevant for years.
First, I don't understand what's the ASF's position here. Do they have some political reason to keep kicking this can down the road? Is there money on the line? Is it just somebody refusing to give up?
And given the licensing allows, why don't they rebrand LibreOffice, rather than keeping a zombie of a project?
Or why don't they just make some sort of deal with LibreOffice?
The pointless commits combined with the lack of transparency feel a bit nefarious. I wonder if it's more about personal drama between the projects than anything else?
And here I thought they were just against OpenOffice, this interpretation pits them against Free Software more generally. I'm gonna run with that for a while and see how it fits.
The members of the OO project want to keep running their project. They have a group of people who keep showing up to do user support, documentation, and produce a huge community around their templates. They just haven't pushed any new features for a while. But they still are there, keeping the lights on.
The ASF is not a top-down organization. It's driven entirely by those projects. And if you look at the dev list - https://lists.apache.org/list.html?dev@openoffice.apache.org - you'll note that this issue gets raised on a VERY regular basis, and the project participants always say "no thanks."
As to "making a deal with LibreOffice", I would encourage you to read the mailing list, where that, too, gets raised with great regularity. There, too, the community sentiment is "no thanks." On both sides.
AOO being moribund has come up at ASF board meetings, and they really did just kick this can down the road.
like c'mon, you're literally an ASF guy, you should know this happened