Hacker News new | comments | ask | show | jobs | submit login
An Open Letter to Apache Foundation and Apache OpenOffice Team (gnome.org)
119 points by mkesper on Aug 18, 2015 | hide | past | web | favorite | 78 comments

For some context, let's look at the read-only github mirrors of each project and see how much work has been done in the last 30 days on both projects.

LibreOffice - core: "Excluding merges, 106 authors have pushed 1,064 commits to master and 4,434 commits to all branches. On master, 4,833 files have changed and there have been 78,361 additions and 54,441 deletions."


apache - openoffice: "Excluding merges, 4 authors have pushed 6 commits to trunk and 8 commits to all branches. On trunk, 16 files have changed and there have been 175 additions and 118 deletions."


tl;dr: OpenOffice had only 8 commits by 4 authors vs LibreOffice's 4,434 commits by 106 authors in the last month

What are all those LO commits doing? I don't think they're adding new features. I compared the two calc codebases recently, and I found no new features in LO. I believe LO has consolidated the number of string classes internally, which is a nice bit of refactoring, but I can't think of any significant feature they've added to the spreadsheet app.

> I don't think they're adding new features.

Wrong, tons of new features land in major releases. Here are the release notes for the feature-packed 5.0, released a few weeks ago: https://wiki.documentfoundation.org/ReleaseNotes/5.0

In addition to this, yes, refactoring is going on: https://people.gnome.org/~michael/blog/2015-08-05-under-the-...

Thanks - that's interesting. It does look like Calc 5.0 has enhanced FLOOR( ) and CEILING( ) worksheet funcs for better MS Excel compatibility.

You often won't see the fruits of massive amounts of cleanup of technical debt, but you might eventually notice one being snappier or less glitchy. Such is the burden of software development.

Or: LO has more commits on a typical weekday than AOO has had in 2015.

Didn't most people switch to LibreOffice anyway?

I mean that's only a loss of 4 authors for LibreOffice.

The problem discussed in the article is not the number of developers on the project, but the fact that, at least to the author of the open letter, users are still installing Open Office because that is what they always did. The author is asking the ASF to redirect requests for OpenOffice to the LibreOffice site so that users can use software that is being actively developed.

Long time linux user. Tend to stay behind the times in terms of distros, 12.04 is still one of my development machines.

I tend not to keep up on the news lists in terms of the politics around stable software that gets the job done. I had no idea IBM pulled the plug on OpenOffice.

For my Mac laptop, I've Open Office installed. Thanks to this open letter I immediately converted to Libre Office. Thanks Christian...

I had no idea IBM pulled the plug on OpenOffice either, because I'm sitting here with my IBM-issued Red Hat laptop and I'm doing project documentation in OpenOffice, which came pre-installed.

I think what they mean is "IBM stopped developing their own fork of OpenOffice (called IBM Lotus Symphony) and switched back to regular OpenOffice".


"Much of OpenOffice's recent decline may be due to IBM's withdrawal from the project. An anonymous informant alleges -- and web searches appear to confirm -- that IBM did nothing to publicize OpenOffice 4.1.1 when it was released on August 21, and that, since then, IBM developers have disappeared from the OpenOffice mailing lists."

Wonder when I will be upgraded to Libre Office then. We've definitely had OS and software updates since 2014...

Are you able to ask for software?

If you read my comment history, you'll find I'm pretty knowledgeable about software in general and FOSS in particular. I only lightly use Office applications, and I had no idea OpenOffice was dead until I saw this thing called LibreOffice in my distro a few years back and was like "WTF happened to my OpenOffice?" and started Googling.

So it's not just a problem for non-technical users -- keeping track of when software you don't use on a daily basis becomes obsolete is a problem faced by everyone.

I heard of this new thing called 'systemd' Have you heard of it? I don't know how I am supposed to keep up....

I didn't even know OO still existed. With that said, I don't find LO much different. It looks and feels the same. The interface is still ugly, like everyone already mentioned, but that doesn't even matter to me. I find the performance sucks compared to Excel. Have you ever tried a multi column sort on a large table in excel vs LO? There's a huge difference. Aside from calc slowness though, the rest of the suite works just fine.

I am surprised the two projects have not merged by now (if not in the sense of a code merge, then at least in terms of people involved).

The OpenOffice/LibreOffice split made sense at the time, IIRC Oracle was being very uncooperative on the matter. But that particular problem has disappeared, and having two projects no longer makes a lot sense.

> then at least in terms of people involved

AOO has so few commits that this has basically happened.

The libre office team should really start a campaign to make everyone know that OpenOffice is dead and Libre Office is the future. The standard user still thinks that OpenOffice is the free office and most of the standard users would never have heard of Libre Office.

Of course it would be helpful to redirect visitors of the OOo website, but I fear this would not happen. So the best thing would be to start a "switch to Libre Office" campaign.

The libre office team should really start a campaign to make everyone know that OpenOffice is dead and Libre Office is the future.

WTF? No they shouldn't. How about both projects just keep moving forward in their own way, and deliver the best results they can, and let people decide what they want to us? There's no need for this kind of public smear campaign conducted between OSS projects.

Better yet, would be more cooperation between the two projects, and a dropping of whatever enmity / bad-blood exists between the two "camps".

I don't want a "smear campaign" but more an "education campaign", it's a matter of the tone. I wouldn't suggest such a campaign if OpenOffice would be alive, but is dead. Dead as a doornail.

We try not to start battles with other open source projects. Live and let live.

There's a lot of FUD in this article. There's no question that AOO development took a hit by the IBM decision, but it takes time to overcome setbacks like that. But I, and a lot of others, believe the world need a nice, full-featured office suite, licensed under a permissive license. People will keep working on AOO, even if not so many as when IBM was contributing.

> People will keep working on AOO, even if not so many as when IBM was contributing.

As noted in the post, this is observably not the case.

See also this article giving the objective numbers: https://lwn.net/Articles/637735/

AOO is a dead project squatting a well-known name, to the detriment of the actual end users.

As noted in the post, this is observably not the case.

No, actually it isn't. If 0 people were contributing to AOO, that would be the case. If 2 or 3 are contributing, then what I said is accurate. As it is, there are still somewhere around a dozen active contributors as best as I can tell. That means the project is substantially diminished, but it's a long way from "dead".

AOO is a dead project squatting a well-known name, to the detriment of the actual end users.

Quit spreading FUD.

OpenOffice had only 8 commits by 4 authors vs LibreOffice's 4,434 commits by 106 authors in the last month

See my above comment: https://news.ycombinator.com/item?id=10078640

How many commits AOO had, versus how many LO had, is irrelevant vis-a-vis the statement "OpenOffice is dead". Nobody is contending that LO isn't the more active project.

With IBM's developers no longer working on OpenOffice, the contention by many is that AOO development is all but dead due to the fact that there aren't enough active development hours going into the project to even maintain the existing codebase (fix high-level crash bugs and critical security issues) let alone make progress on any new features or standard bug fixes. The extremely low volume of commits in AOO seems to match up with that contention. The comparison to LibreOffice is relevant as it is a comparison of a healthy open source project with multiple contributors to one which is said to be dying. LibreOffice has more commits on an average weekday than AOO has had in all of 2015.

The trouble is that they've already ceased their developer activity. That’s the entire point of Schaller's post.

Look at what they're doing. They are serving up a Windows installer that is known to have a serious vulnerability: anyone can send an exploit in a HWP document with a .doc file extension, and an AOO user opening it will be pwned. The AOO "devs" know this is the case.

Everyone downloading AOO 4.1.1 between April 2015 and August 2015, they have done them actual harm and installed a vulnerability on their machine.

The fix is trivial: repackage the Windows installer minus one file.

Instead of doing that one simple thing, they're posting excuses for not doing it.

(If the answer is something like “well, Apache procedures make it difficult to just rerelease an installer minus one file”, then this would not be evidence in their favour - it would be evidence that the AOO bureaucracy actively hampers the interests of users and is not fit for purpose, and that Schaller's blog post is entirely correct.)

You can’t expect people to take seriously claims that the project is not dead when they're actively harming their users by serving them software with a serious vulnerability for the past four months, and spending their time making excuses instead of fixing it.

First project to add the missing "Outline View" functionality gets my eternal loyalty :) https://bz.apache.org/ooo/show_bug.cgi?id=3959 https://bugs.documentfoundation.org//show_bug.cgi?id=38093

This RFE is a healthy young teenager now and I fully expect it to outlive me. Happily I don't have so much of a need for word processing these days.

It's basically waiting for someone to bother writing it.

Why not create a bounty then?


Or pay one of the certified LO developers to implement it:https://www.documentfoundation.org/certification/developers/

I sponsored, loading Adobe colour palettes in 5.0 myself.

I thought there was a bounty for it on FreedomSponsors already, though it appears there isn't one yet ...

Glad this letter was linked here. Despite being technical I had no idea that OpenOffice was in death throws. I'll switch to LibreOffice immediately on my MBP.

I think part of the reason I'm out of date on office suites is that I don't use them often. At work, I'm mandated to use MS Office. When I have a choice I switch between Google Docs and iCloud's office suite. Still, it's helpful to have an office suite on my laptop when I can't access a WiFi connection.

The author's point about letting OpenOffice be an important part of history is a salient one. After MS won the "Wordperfect v. Word" and "Lotus 1-2-3 v. Excel" wars, we didn't have a choice other than purchasing MS Office for a long time. For light users like me, it's not worth it to purchase an office suite I use once every few weeks. OpenOffice and their ilk has been important for people like me. (EDIT: grammar)

Star Office, Corel Office, Lotus Smart Suite, Appleworks....

Corel Office included Borland's Quattro Pro, and WordPect, which Borland bought then sold Novell, which sold it to Corel. It seems Corel Wordperfect Office X3 - Home & Student Edition (PC) was launched in 2007. (1)

Lotus Smartsuite Millennium 9.0 was launched in 2008, including 1-2-3 and a bunch of bought-in programs. IBM dumped lots of copies on the market by bundling them with PCs, as part of its attempt to kill Microsoft. (2)

They probably got killed off by OpenOffice....

(1) http://www.amazon.co.uk/Corel-Wordperfect-Office-X3-Student/...

(2) http://www.amazon.co.uk/Lotus-Smartsuite-Millennium-9-0-x/dp...

OpenOffice was (long ago) based on the Star Office code base... Which was then forked to Libre Office after Oracle purchased Sun.


True, but Sun continued to sell Star Office with its own additions. I believe it sold it directly with commercial backing as an alternative to Microsoft Office.

Oracle even released a version of Oracle Open Office before giving its code and trademarks to the Apache Software Foundation.

It's interesting to think about this in the long run rather than the short term. Perhaps given enough time the projects will diverge to something significant? Sort of like all the variants of Unix.

You are right, but for that to happen, AOO should ship code, which is not doing at the moment, AFAIK.

They already have: LO has added functionality (e.g. embedding an .odt in a PDF file, effectively making the PDF file editable or font-embedding), which AOO simply does not support.

As good as all features listed in LO release notes are LO only.

Does LibreOffice have better MS Word compatibility? It only happens once in a blue moon, but my wife will receive a word document from someone (most recently a resume that she had professionally written) in Word format, but OpenOffice breaks it completely when she just wants to go in and change one sentence. Having to deal with word documents only once or twice a year obviously isn't enough justification to pay a subscription for Office 365, but are there any other alternatives with maximum compatibility as a strong priority?

If your primary concern is "maximum compatibility" and you only need to use it "once or twice a year" wouldn't Office Online[1] be sufficient to your needs?

[N.B. - I've never used it, just seems it would suit your use case]

[1] https://products.office.com/en-us/office-online/documents-sp...

Vastly better. OOXML is a bit of a Zeno tarpit http://esr.ibiblio.org/?p=6772 but LO is significantly better every release.

(And vastly superior to Google Docs, which is simply incompetent with really quite mundane .docx files.)

You probably won't get perfection, but LO is already more compatible than MS Word is with itself between versions.

Yes. The RTF filter has been improved a lot and this the doc and docx filters did too (these filters share a lot of code).

Unfortunately it's very depressing reading, largely 'my 0.0001% problem shows that the new stuff is bad', and 'I subjective opinion therefore x interface is bad'.

The sad thing is, they're completely oblivious of the sidebar. The idea behind it is, to use the much larger vertical screen space of current monitors. In 5.0, you can hide all toolbars and exclusively use the sidebar.

Try it!

Someone in there mentioned trying to make a "WordPerfect killer," completely oblivious to the fact that WordPerfect is dead and Word is what killed it.

No need for Microsoft help, they were quite good killing themselves.

There is even a book from one of the founders about it.

Where is that book? I would like to read it.


If your goal is migrating users from MS office you won't have much chance but offering something they now have become used to.

I am a software developer and "technical user". I don't care about new features. I tried switching from OO to LO and my experience was that it is full of bugs and not 100% compatible with my existing corpus of OO diles. I'll be sticking with OO thank you.

Why is it that anytime LibreOffice or OpenOffice is discussed, it takes about 3 comments before somebody starts talking about Microsoft's ribbon interface? How is it still a contentious subject?

I think I probably read those comment threads because it makes me feel like some kind of genius. I have the seemingly rare ability to use either the LibreOffice menu+toolbar interface or the Office ribbon interface.

That said, I don't like using LibreOffice because it's ugly (IMHO of course) and beautiful things work better.

LibreOffice is ugly, but a fair amount of the ugliness comes from the effort to shoehorn more and more features into the same style of UI design that inspired Microsoft to develop the Ribbon UI to begin with. Yes, I suppose it's possible that LibreOffice could come up with their own, more modern UI. But I doubt it. UI design has never been a strength of theirs, and the Ribbon UI comes from an incredible amount of research at Microsoft that costs rather a lot of money to do. So the question of which style of Office UI to use is eternally a question - people who hate the Ribbon will insist on holding out, but it makes LibreOffice look more and more dated the longer they do.

Some of the ugliness came from implementation decisions that LibreOffice have been busy fixing.

For example the dialogues used to have to be laid out in such a way that all languages strings would fit in the space that was left. Which left them a bit sparse and odd looking. I think that conversion is all done now:


Wow - what a difference that makes. It's very encouraging to see visuals getting some attention.

It would have been nice if Microsoft had at least made the ribbon bar optional and allowed people to revert to the old interface. The whole debate would disappear if you could choose the UI you prefer and be done with it.

That being said, I was mildly surprised that most people I have spoken to about this seemed to like the ribbon bar, especially non-techies (who are, after all, the main audience for MS Office).

There's a cost to maintaining old interfaces, and the old UI had only one advantage: some people already knew it well and weren't prepared to change. However, they could still use their old version of Office, simply by not upgrading.

> most people I have spoken to about this seemed to like the ribbon bar, especially non-techies

I'm a techie and I love it. The old UI now looks incredibly clunky. What really surprises me is that some people are still thinking about a change that occurred in 2007 ;-)

admittedly, the new ribbon interface (well I guess it's no longer new, it's just that I went from using excel for about 6hrs/day to maybe 12hrs/year circa 2005) does look prettier -- but how do people actually go about finding anything in it? does everyone just use google & try to find some blog post with screenshots?

> but how do people actually go about finding anything in it?

The discoverability is pretty much the same as the old pull-downs and toolbar approach (and how you find things is pretty much the same -- click on the category whose label seems like what you might, and then look through whats revealed by that), but consolidated into what amounts to a set of toolbars organized into tabs, rather than a combination of toolbars and menus.

(And what often gets ignores is that a lot of the improvements have been in the contextual right-click pop-up, which means what you usually need often doesn't require using the ribbon at all.)

> does everyone just use google & try to find some blog post with screenshots?

No more so than with the old interface.

> The discoverability is pretty much the same as the old pull-downs and toolbar approach

One of the main points -- possibly the main point -- of the move to the ribbon was to increase discoverability, and it worked. Gates said Microsoft had lots of compliments about new features in office 2007 that had actually been in the old version, but people hadn't found them.

Microsoft did an enormous amount of work (and research) when developing the ribbon. Jensen Harris covered this in great depth in an 8-part series at http://blogs.msdn.com/b/jensenh/archive/tags/why+the+new+ui_...

I'm pretty sure you can find videos of him giving presentations about it as well. (I saw him present.)

>It would have been nice if Microsoft had at least made the ribbon bar optional and allowed people to revert to the old interface. The whole debate would disappear if you could choose the UI you prefer and be done with it.

Don't underestimate the cost of having to support two sets of UIs and to provide training for that as well. Ultimately a unified UI works best, at least for organizations, which are Microsoft's biggest customers.

All of the Openoffice/LibreOffice folks that I know are generally very.... passionate about the ribbon.

My understanding is that ribbon was not considered for LibO due to the patent controversy: https://en.wikipedia.org/wiki/Ribbon_%28computing%29#Patent_...

The design team is focusing on improving the sidebar. It does make sense as the typical displays now have much more screen estate in the x axis. Calligra is another office suite with a sidebar.

Btw. the team is always looking for new contributors: https://wiki.documentfoundation.org/Design/Team

"beautiful things work better"

Beautiful things appear to work better; less cognitive dissonance. The world of project management is replete with instances of software being replaced by something that looks better only to find that it doesn't actually work better (and the converse as well).

In the world of the arts truth and beauty are typically correlated, other realms of human endeavour less so.

It really helps out in life in general if one assumes that the author meant "all else being equal" unless the author explicitly contradicts it.

I would not say that LibreOffice/OpenOffice is ugly. It is not very pretty, but I think that is okay - I use it to get work done.

I can live with ribbon interfaces, but I still think they were a bad idea, made all the more annoying by everyone and their grandma jumping on that particular train. But, for better or worse, it is a little late for this discussion.

Count the number of widgets used in the top two inches of screen real estate. If it is greater than say 10, that isn't beautiful; not for someone trying to figure out how to use it.

how is it Apache's responsibility? how about an open letter to the LibreOffice team to do what every single commercial company does, go out there and sell itself. OSS work is often noted and praised for having evolutionary and organic growth characteristics, as opposed to the coordinated and centralized growth exhibited by MS Office. without exhausting that vein of thought, this open letter seems to want to have it both ways.

If you had a straight battle for name recognition between LibreOffice and Apache Open Office, which would win?

AOO is just benefiting from years of hype behind OO.org, which would be fine if it had community support and was making the same sort of improvements to the code....

LO forked before AOO existed, so you could argue that https://www.openoffice.org/ should offer to link to both versions. (I'm not wholly convinced by this myself, but it would level the playing field, and more people might download the better version.)

and,,, you seem to have completely have missed the point.

if OSS is organic and evolutionary, we shouldn't be discussing market-share strategies. LO, keep on doin'. AOO, you keep on doin' too.

i wouldn't argue anything particular here, just would like to point out that the sentiment of the open-letter is one of coordinated community action, which emphasis on the coordinated.

if i wanted coordination, i'd join an organization, like, maybe, a company, like, maybe, microsoft. naw, i'll just keep on doin'.

Not really. The whole point of Schaller's Open Letter is that he thinks people are being led to use an inferior version of open source code. In fact, he says: “Letting users believe that OpenOffice is still alive and evolving is only damaging the general reputation of open source Office software among non-technical users.”

This really doesn't have anything to do with market share per se.

He just thinks that marketing an inferior product with almost no community support is bad for users and bad for open source.

The Document Foundation is not a company! It's a foundation!

In my world this is mostly irrelevant. I use GDocs for word processing and spreadsheets. For presentations it's Prezi. (Disclaimer: I work for Prezi.)

Interesting. I too use GDocs and Prezi sometimes, even though I usually try to use Free Software.

After googling a bit, I found some open source projects that let you host your own web office solution. It would be hard to get friends and coworkers to switch, though.

I wish someone would do for libreoffice something like firefox was for mozilla, creating a new product out of the codebase with better UX.

We updated the link from https://lwn.net/Articles/654776/, which points to this.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact