Why is Apache clinging to OpenOffice? | Hacker News Reader

Why is Apache clinging to OpenOffice?

arencambre.com

379 pointsAuthorizedCust355 comments

The worst thing about this was that openoffice was starting to get some name recognition before oracle killed it, and this whole debacle intentionally or unintentionally turned that name recognition into a weapon against LibreOffice.

Ideally they should have given LibreOffice the OpenOffice name, but even if they had just said "openoffice is dead" then it would have been easy for people to find libreoffice.

Unfortunately, by taking over the OpenOffice name and absolutely refusing to admit that development has ceased, Apache has caused tons of confusion that has massively harmed libreoffice, because for years people who had heard about "openoffice" would see the terrible, out of date, insecure pile of junk being distributed by the Apache Foundation and not realize that the actively developed LibreOffice was available.

In the case of linux distributions it was less of an issue since they just switched to LibreOffice, but I wouldn't be suprised if there are windows users who are still downloading OpenOffice (and possibly taking one look at it and giving up and switching to Office 365).

Apache should have killed OpenOffice years ago but at the very least they should take it down and honestly apologize for fucking this up so badly.

Even ignoring the harm to LibreOffice it's just irresponsible for Apache to keep distributing OpenOffice when they can't even make new releases to simply apply security patches for years.

Yeah, that was me, I totally downloaded it as the first thing when setting up a new computer for years. I thought the errors opening some excel documents was because it was open source. Little did I know it was effectively abandonware.

If you let the project die on the vine however, you can clean up after it dies without having to take the hit for actually being the ones to kill it.

Apache is in an unenviable position, here. Might as well let nature take its course.

I really don't understand why Apache doesn't just put the OpenOffice branding on LibreOffice. Is there any real reason to not just make OpenOffice a branded distribution of LibreOffice at this point?

edit: oh right, licenses. LibreOffice is MPL, and the Apache Foundation won't touch it.

My guess would be corporate bureaucracy

I like name LibreOffice much better.

Maybe because when I tried using OpenOffice while I was at university it sucked really bad.

I tried LibreOffice like 5 years ago and I am using it day to day and I donated money to them and consider donating again.

Yeah, when I hear Libre Office, I think "That FOSS word processor"

When I hear Open Office, I think "That dystopian floor plan corporations use when they decide cubicles aren't soul-crushing enough"

It might be that - Libre seems to sound or feel more like "free". Where "Open" feels like corporate BS, maybe even worse like corporate BS that some crushed soul came up with.

Meh, I guess it's all subjective, but OpenOffice has a certain alliterative, authoritative-sounding ring to it. The name LibreOffice feels clunky & hokey in comparison. It's a bummer that the latter wasn't able to adopt the branding of the former.

Agreed, also "libre" is a gnu/jargon word which doesn't exist in most people's vocabulary, and probably seems meaningless or foreign.

> in most people's vocabulary

In (monolingual) English speakers vocabulary.

As a French native speaker I find it even weirder. How is "libre" even pronounced in an English name?

Sort of like Libra but with less emphasis on the second vowel.

Definitely no French R.

I pronounce it as something like /'liːbreɪ/ (LEE-bray). (It definitely feels to me like a conscious foreignism.)

My mom's German. I grew up in a bilingual household.

It's not a big deal to my mind. (shrug)

According to my dictionary, the pronunciation of "libretto" (an imported, but already established English word) in IPA is [lɪbr'etoʊ]. We can extrapolate from this.

Wouldn't that lead us to pronounce it as "lib-RAW-fuss" instead of "LEE-bray-off-fuss"? Not that it wouldn't be funny.

It would lead to that pronunciation as English likes its schwas and disappearing schwas, just like French does. (Someone else pointed out that "swallowed" schwa is closer to the French pronunciation, too.)

I've lived in parts of the country where "libre" and its counterpart "gratis" were somewhat more common English vocabulary words, and their pronunciations were much more in line with "rhymes with zebra" (LEEB-ruh) and "rhymes with lattice" (GRAH-tuhs). To be fair, those were parts of the country with a bit of a "South midwestern" drawl and some might take those to be lower class, "less educated" pronunciations, but they certainly sounded a lot more like the natural pronunciations of such words in English to me.

All I’ve heard in Australia (from only a very few outside my family) for LibreOffice is /'lɪbriː/, LI-bree.

In this case, I hear it as "lihb-ray-office" from random users.

Most US English speakers will pronounce it the Spanish way.

As in Spanish.

leeb-ray

Most Americans at least heard of lucha libre, so the Spanish pronunciation is familiar.

Even for monolingual English speakers, like myself, it only takes two or three braincells to make a guess that "libre" comes from a Romance language and probably relates to the word liberty. That gets you close enough to understand the sentiment behind the name LibreOffice.

I don't think the Libre name is a problem at all. It's just one of those things people too cynically think will be a problem for other people. The real problem for LibreOffice was losing all the brand recognition OpenOffice managed to accrue in the years before everything went to shit.

I don't quite agree. It's not so much the meaning as the ease of use that's the problem. You don't need to have any worries about pronouncing, nor spelling, OpenOffice correctly. You can tell somebody about it without going through the whole "Huh? That's how you say it? Are you sure that's right? How do you spell it again?" routine.

I'm not saying it's the whole story, or even the biggest part of it. But it adds a bit of friction to getting the word around about the new name. See uBlock Origin for a more successful example, where the mere mention of uBlock had, and still does to an extent, "you mean Origin, right?" as a default, low-effort response.

"Lee-bray"

I don't know if the French would approve of that pronunciation, but it's good enough for me and seems like an intuitively obvious way for me to pronounce it. The word looks obviously French, and monolingual English speakers encounter French-looking words often enough that it doesn't give many people pause. Just pronounce it like it was English and put on a fake French accent if you want to get fancy with it.

I won't say there is zero friction from the word Libre, but it's certainly much smaller than losing all the built-up brand recognition of OpenOffice, and even that is much smaller than the elephant in the room: the ubiquity of Microsoft Office and the trouble any competitor will have no matter what name is used.

I think people fixate on the name because that's the easiest thing to change. But changing the name would only make things worse, by once again wiping out all accrued brand recognition. Even if you ignore that and want to talk about a hypothetical situation where it was named something else originally, the project would obviously still be an obscure oddity in the global Microsoft monoculture.

> I don't know if the French would approve of that pronunciation, but it's good enough for me and seems like an intuitively obvious way for me to pronounce it.

I say it the same way.

Here's the catch: if you pronounce it at someone as a recommendation i.e. "you should try Libreoffice!", how likely are they to remember it when they get home, or to be able to spell it to search for it?

You do know there is also a Spanish language, where "Lee-bray" is pretty close for an English speaker.

> mere mention of uBlock had, and still does to an extent, "you mean Origin, right?" as a default, low-effort response.

This is frustrating but important, because uBlock not-Origin is a parasitical scam with good SEO.

Hell, at the very least it makes me think of Nacho Libre - and who doesn't like Jack Black?

> and probably seems [..] foreign

Yes, it's French for "free" after all.

It's free in Spanish too. The word has a Latin root.

"Why did you put this French stuff on my computer instead of All-American™ Microsoft?"

LibertyOffice?

Absolutely better. OpenProductivity? When I ddg it, the third hit is OpenOffice, so they're probably buying it, it's so good. CornerOffice?

This right here. I don't know many regular folk that ever use the word Libre, but many use the word open and office.

I just think of the Nacho Libre movie with Jack Black.

That part of the open source community literally prioritises linguistic whimsy over success.

In that particular case the openoffice name was trademarked so they wanted to find something bearing a semantic ressemblance.

What was your suggestion for a name and where were you when it was time to choose it? ?

> bearing a semantic ressemblance

But it doesn’t, outside of the Free Software fandom. They picked a name for themselves, not for their audience.

The name sucks, but libre absolutely does bear a rather non-trivial semantic resemblence to open.

Come on.

Not to mention Libre being a synonym for Free Software (the GNU kind).

Approximately zero people understand that connection in the general population. They're screaming at the choir.

That is not true

Like 1 Billion people in the world are native spanish or french speakers and understand libre verbatim. Add to that most people speaking a latin language will also understand it despite the slight difference in spelling. Most people speaking english know the word liberty and those that have an IQ higher than an oyster do figure it out easily.

Additionally you don't have to have a connection for a name to work. I mean there are a lot of software used by millions of people without them never ever thinking why they are called like that.

People in this thread, so technical people, who already have an interest in OpenOffice / LibreOffice, are saying they don't even know how to pronounce it!

I still ain't seeing your suggestion for a better name.

Asking people like me is how you get a crap 'developer art' name like the one they have - they should have asked a professional!

I guess you were already a generous donator back then already then!

shoulda woulda

I think crowdsourcing a name by polling the public will get you something like OfficeMcOfficeFace, and that is if you are lucky. So if this is your roundabout way of pointing out that the name could be even worse, well played.

I mean, at least OfficeMcOfficeFace is memorable and straightforward to pronounce.

Yup. When I hear OpenOffice I think of something professional. When I hear "libreOffice" I think of the Adwattia icon set and gnome crashing because of some broken default.

I don't know why you're downvoted. LibreOffice is the superior product and perfectly good, but this is a perfect expression of the mental image the two names evoke.

The actual quality of the software is a seperate issue. I didn't interpret your comment as attacking the software. The office sofware anyway! :)

One of the problems with OpenOffice was/is that it's a trademark in Brazil and the Netherlands. I don't know much about the Brazilian case, but in the Dutch case it's a small IT consultancy focusing on Linux and other Open Source stuff. Especially given they operate in the same kind of space that OpenOffice.org (the office suite) it's led to some amount of confusion.

LibreOffice sounds like a co-working space founded by college libertarians.

I am not really a hipster type - but still like that name :)

I like name LibreOffice much better.

"FreeOffice" would have been better, if available. If you want mass adoption, use a word the masses use.

Apple doesn't spend billions on branding because it doesn't work.

There is indeed a product with the name FreeOffice, in the same category, by SoftMaker.

Maybe call it something other than "X"Office? The names are too few and too generic to be a meaningful distinguisher, IMO. It's like how every hobby assembly is <letter>asm. Slapping "free/open/libre" in front of a generic word has limits.

Freeoffice as a name is a bit reminiscent of the junkware you used to get on discs with magazines.

I wonder what the modern equivalent is, probably the websites that show up in results if you google for US free tax software or video conversion tools that are just bad ad-filled guis around ffmpeg probably.

Many of the people using it are not old enough to remember that.

And I would wager many of the people in that group of not being old enough to remember it are not in the market for any desktop office suite.

> I like name LibreOffice much better.

There's no accounting for taste, but if past threads on the topic are to be believed, there is not even a consensus on how it's pronounced. That alone makes it a highly questionable product name.

I can only imagine this is split between those who (correctly) think it's pronounced "Lee-bra"[1] (like Libra the star sign) or "Lie-bra"[1] (like fibre).

I can't imagine how on earth it would be possible to be so unaware of other languages that one could even arrive at the second option.

[1] In both cases the A is a "short" A.

"Lee-broffice".

This is is how I say it, knowing it's technically wrong and sloppy but my convenience trumps that.

Clearly the outcome anyone wants for their name. Nailed it.

Re-fork it to just "Broffice". $$$

As has been mentioned, libre is also a Spanish word, and an anglicised pronunciation derived from Spanish would be something like 'lee-bray' (although of course the last vowel isn't a diphthong in Spanish).

You've found a pronunciation that I've never heard a single person use (your confidently "correct" one.) The second one, "lee-burr", I've heard. "Lee-bray" is nearly correct if you think it's from Spanish, and "Lee-bruh" is nearly correct if you think it's from French, where the "a" is a schwa (ə), not a "short a."

I had forgotten about Spanish pronunciation (more like "Lee-br-eh" than "Lee-br-ay" if memory serves correctly, but was about 35 years since doing a couple of years of Spanish in school so I won't be nearly as over-confident).

I was thinking of French and meant "the "a" is a schwa (ə)" when erroneously calling it a "short a", I never know the correct names for the different vowel sounds.

I'm really not sure how you arrived at "a pronunciation that I've never heard a single person use". I may have used wildly incorrect terminology, but given the star sign example I'm curious how you pronounce Libra that is so wildly different from "Lee-bruh" (being a reasonable approximation of French "libre")?

I pronounce it "lee-bray" office, but that's probably my southern US roots. lol

Speaking purely based on the name, not the actual software:

What comes to mind when I hear the name LibreOffice is historically mediocre FOSS that you'd only choose for ideological reasons, which is why the libre part needs to be advertised up front. When I imagine what it looks like, I'm imagining some creaky UI that looks like GIMP from 1999. It's worse, even, than OpenOffice, because OpenOffice is the obvious name if you're going for the ideological angle, which suggests that LibreOffice is a shitty knockoff of a shitty knockoff. Honestly even StarOffice is more appealing to my ears.

I'm positive that all of this is deeply unfair, but that's what comes to mind. Part of the negativity probably relates to conflating LibreOffice and OpenOffice in my head.

Now when I will hear libreoffice again, what will come to mind are clueless illiterate gringos ridiculing themselves on hacker news.

> because OpenOffice is the obvious name if you're going for the ideological angle

I disagree. I associate "Open" with corporate-interest driven "open source", BSD licensing, and cynically misleading branding (OpenDNS). 'Open' is the word you use if specifically don't want to invoke GNUy left-adjacent ideologies.

Open Server.

But even with these damaging examples which I agree with you about, it's still better.

> StarOffice is more appealing to my ears

Now THAT's a name I haven't heard in a while. And it does sound vaguely attractive, certainly more than the two alternatives.

My embarrassing mistake is that I don't accidentally call Libreoffice Openoffice as often as I accidentally call it Staroffice.

Someone should sue them, for knowingly, willlingly releasing bug ridden, security issue laden software.

Seriously.

This is a perfect test case.

It should start with a 15 day demand letter, and progress from there.

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.

A warranty is not the same as willful negligence.

Why is there no resposnibility on the part of the user?

The software equivalent of “shall not be infringed”.

I see downvotes, but there is a difference between "our stuff has a bug" and "we're actively keeping this project alive, including updating binary downloads, even though we know it has endless CVEs, some of them years old."

One is an innocent problem. The other is willful negligence.

And we need to start suing for this sort of thing. We need fines for companies willingly causing harm.

> we need fines for companies

I think the reason you’re getting downvotes is that Apache is a non profit foundation, not a for profit company. So fining them isn’t going to do a lot of good (as well as being very unlikely to succeed)

> Apache is a non profit foundation, not a for profit company. So fining them isn’t going to do a lot of good

Why?

I don't see why fining non-profits is any less effective than fining for-profits.

It might be effective... but is it doing any good?

If you think getting OpenOffice off people's machines is good.

Possibly even more effective.

The point is to get them to fix, or take it down.

A fine per day, until compliance, can motivate well.

Being nonprofit doesn't mean you can willingly hurt people (getting hacked) either.

A fine for what crime?

Nonprofits can obey a court orders same as anybody else. When they are obviously in the wrong, it is hard to understand why they would resist.

That said, it is hard to understand why they resist good sense now.

Except, "Apache is where free software projects go to die."

> And we need to start suing for this sort of thing.

Sue them... for what? What do you plan to put in your paperwork? What are you going to say to the judge?

Knowingly distributing a faulty product with risk of harm to the user ?

I think you’d have at the very very least specify an actual harm against you, and even then you’d likely be told immediately that they have no obligation to provide anything given there’s no support contract.

How would you define "faulty" and "harm"?

rm will unapologetically delete files instead of using the "trash bin" semantics that many people are used to. Some would define that as "faulty", and it can certainly cause "harm" (a "rm fuckup" is almost a rite of passage).

You can find many such almost banal examples, ranging from well-known tools to some project a student uploaded on GitHub that sees basically 0 traffic. Opening up Open Office to a lawsuit also means opening up countless GitHub projects from 15-year olds riddled with SQL injections and the like, but also things I put on my GitHub five years ago and don't really care about. Ignoring a PR would mean risking a lawsuit.

Plus, do we really want government involved in telling us what software we can and can't put on the internet? Because that's what this would mean.

"They should be sued for distributing outdated insecure software" is a fun one-liner, but the ramifications if it would actually happen are huge and almost entirely negative for the Open Source world.

IANAL, but the license includes both a a disclaimer of warranty and a limitation of liability. Further I think to sue you need to demonstrate actual harm. Absent a security breach, for which you might get a day in court, they haven't harmed you.

Stop using their software. Stop bullying people with lawyers.

> This is a perfect test case.

Indeed, and maybe your legal theory fails.

There are plenty of jurisdictions where such a disclaimer is void. France, for example, probably. Or anyway somewhere in Europe.

News that Apache had been forbidden to perform downloads to France, and then more places, might help raise awareness that it shouldn't be downloaded at all, nohow.

Maybe it would even push Apache to stop offering it entirely, and to link to the Libreoffice site instead.

> There are plenty of jurisdictions where such a disclaimer is void. France, for example, probably

I would be surprised, as such disclaimers are included in the official French flavor of the GPL family (http://cecill.info/licences.fr.html); section 8.2.

I would assume that the disclaimer protects the copyright holder but not the distributor.

The European approach to consumer protections is generally more about ensuring that things are safe by default than about warnings and disclaimers. A lot depends on the expected competence of the target audience. If you distribute professional tools directly to professionals, they can be expected to read the warnings and understand that misuse could be dangerous. If you market and distribute something to the general public, ensuring safety is your responsibility.

> I would assume that the disclaimer protects the copyright holder but not the distributor.

Why assume when you could check? Because the license text shows that you're wrong, just read section 8.2 once again.

Because it's a matter of law. Legal protections often override licenses and other contracts.

Well, the CeCILL licenses have been validated by the French institutions themselves, I hope they know their own laws better than you do.

So, someone should put all of the free and open source software at risk (if the "we guarantee nothing" part of their license doesn't hold ground for them it holds ground for no one) because you want to attack the ASF for not doing what you want? What a great plan.

I know people who even nowadays install OpenOffice and I have to explain them that it's not what they should be using, etc, etc.

I just learned this right now, by reading this.

They need to switch OpenOffice to LibreOffice and use both names for the same product (which is called today LibreOffice).

> I wouldn't be suprised if there are windows users who are still downloading OpenOffice

I wouldn't be surprised if as many or more Windows users were downloading it than downloading Libreoffice. The problem is that the name isn't punchy and is unpronounceable in English. Plenty of people who are completely knowledgeable about the situation still refer to Libreoffice as Openoffice.

It's intentional sabotage, more than probably covertly initiated and maintained by Microsoft, and destroyed the entire Apache Foundation for me. I avoid all of their projects, because if that project has a commercial competitor with deep pockets, I can't trust it not to be intentionally bad or broken.

...I'm a native english speaker and I've never had trouble pronouncing libre, is it a problem for others? I've always said "lib-ray", which for all I know is wrong, I never gave it a second thought.

That's a perfectly fine pronunciation, but the point is that it's not an English word, so plenty of people will think it looks weird or not know how to pronounce it.

What till you hear how many non English words are in your everyday vocabulary.

Words in my everyday vocabulary are English words by definition, regardless of whether they're of foreign origin, in the sense that I meant.

"Libre" isn't an English word at all; not even as a loanword from a foreign language.

"Lee-bray" is closer to accurate.

I think "Lee-breh" might be closer to accurate.

FTR, the accurate French pronunciation would be something like leeb-roffice, the "e" in "libre" being absorbed by the o.

But in any case, I agree with GP that the naming is meh.

As a native english speaker, that seems a really much better way to say that awkward combination of syllables.

I see! French pronunciation omits the 'e': https://en.wiktionary.org/wiki/libre

I was thinking of the Spanish pronunciation.

Probably, I'm just doing a bastardized Spanish pronunciation.

"unpronounceable" is technically hyperbole, but the point is true. It's just an annoying speedbump both reading and speaking. Actually more so in writing than speaking.

Speaking, I don't bother to even try to be technically correct, I just say leebruhoffice. It's honestly about exactly as fast and easy as openoffice, but 1, it's only made equivalent by being wrong and sloppy. 2 still the mental process of reading it is just, more work, somehow.

The name just isn't as good at what good names are good at. It's ok. It's accurate. But it's not good. It's just annoying to crank out and stumble over in a way that neither StarOffice or OpenOffice are. NeoOffice is good on those fronts but is bad simply because the word neo is too I don't know, stupid. I can't take anything seriously that calls itself neosomething. "Drive the new neorunner..." pass.

yeah, plus if you tell someone to "download libreoffice", there's a good chance the thing they're going to type into google is going to be something misspelled.

having watched Nacho Libre (Jack Black's "sport" comedy) many years ago, and laughing hysterically all the way through, I have no problems with the Libre part.

But it ends in libre. You aren't trying to go from libre to O-something.

It needs a consonant of some sort. Maybe thats why oopen office works is because it's going from an N to the O.

I couldn't put my finger on that untill you just said that. Part of the problem is definitely that that transition is just awkward.

Maybe they can take one of the excess ducks from duckduckgo. LibreDucks even works better than libreoffice.

And I'm all the way with you on the movie.

I usually pronounce it like 'lee-bur', though at times I've tried a more French version which is closer to 'lee-bruh' but it doesn't flow nicely into 'office'. Hearing 'lib-ray/lee-bray'/'lib-er' (like liberty) I would know what you mean though.

This just means it has ambiguous pronunciation, and there are many things like that. It's not "unpronounceable". Something unpronounceable to an English speaker would be something like words using a non-latin character set or words using latin characters in long or unfamiliar ways like random Finnish words or some words resulting from trying to romanize another language.

Just FYI, in French, we would just skip the "e" and say something like leeb-roffice.

That rolls off the tongue quite nicely in French.

I'm a native French speaker and when speaking English, I just say "Office" because LibreOffice sounds weird. And for similar reasons, I never say "I used The Gimp". I do very much appreciate both projects. Naming is hard.

the gimp is just exceptionally awful, and was apparently intentionally named after the gimp from pulp fiction, which makes it even worse

> I never say "I used The Gimp".

Have you seen Pulp Fiction?

https://www.youtube.com/watch?v=B34DmsMxUlA

we had some people at work running a copy of librespeed, and they kept pronouncing libre as LEE-BER I had to correct them once a week for 2 months before they got it.

If Linux has taught me anything, when spelled all lower case like that, that is clearly lib-respeed

I have that problem sometimes looking at the name libreoffice, too. It's the Reoffice library, clearly

Great, now I can't unsee this.

Thank you for this! Now I have a reasonable way to pronounce it.

> "lib-ray", which for all I know is wrong,

That's fine, but if you're curious, the Spanish pronounciation is "li"- as in "list" or "ligament" and "bre" as in "Brenda" or "bred".

> but if you're curious, the Spanish pronounciation is "li"- as in "list" or "ligament"

This is... not correct. Spanish pronunciation would be LEE-bray. The "li" is similar to "bLEEd". And for proper style points make sure to flip the "r".

It's not bray. You would stop on bra without the y

You're half right. The "bre" e is pronounced like pen or bed, not bray.

And here I thought the schwa just wreaked havoc in English.

Outside of my home country of New Zealand, I think very few English speakers pronounce either "pen" or "bed" with a schwa. (actually, even we don't)

No, not "bray", just "bre" as in "bred" (from OP example).

Native Spanish speaker here.

I'm honestly having a hard time listening to the difference between English "li" and "lee". If you mean how "lee" tends to hold the vowel for longer, maybe you're indicating that that syllable holds the phonetic accent in "libre", which you'd be right about. Fundamentally though, they sound the same.

Not sure what you mean by flipping the "r".

To me (another native english speaker) it's not that pronouncing "libre" is hard, but that then combining it with "office" makes for a clunky awkward mouthful. Whereas "open" and "office" flowed together quite well.

> It's intentional sabotage, more than probably covertly initiated and maintained by Microsoft

Do you have any evidence for this? I would like to think that HN is above making these kinds of totally unfounded accusations.

(I'm not affiliated with any party here and don't care about the conflict at all.)

No evidence, but that doesn't mean that they're unfounded, that means that they're unproven. There's only one party that profits from the confusion, and it's a party that spent a fortune on bribes (or "grants" or whatever) during the OOXML debacle.

Sort of like making an unfounded accusation against Microsoft that can't be disproven. C'mon, we are all better than that here.

The commenter is right to the point. Look at the sponsors of Apache foundation, and you'll see that Google and Microsoft appear as some of the largest contributers. Of course they don't want to antagonize these companies.

The best way to look at Apache and simular foundations is as infrastructure maintained by big companies to redirect open source effort to their goals. They use third parties to do this because the companies themselves enjoy very little trust from the open source community.

So, no evidence. Just FUD. Got it.

The Apache Foundation has next to zero input into how any of the projects are run, so even if this were true it would be irrelevant. How exactly do you imagine this influence working? It would have to be quite overt to have any effect on so many projects that are self organising.

They have 100% control over what projects they office.

I just read it as "lib-reoffice" cause it's impossible not to if you've done manual package management in the past :-D

> I wouldn't be suprised if there are windows users who are still downloading OpenOffice

It gets about 235,000 downloads/week, most of which are for Windows.[1] It doesn't seem SourceForge shows the absolute number of downloads, and while ~235k/week isn't a huge huge number, it's still pretty significant. Dunno how this compares to LibreOffice or other software packages (e.g. Firefox).

[1]: https://sourceforge.net/projects/openofficeorg.mirror/files/

It's a pretty darn high number.

Compound interest on that negative goodwill...

You can click on the chart symbol next to the number

Ah, that's very non-obvious!

Summing up the daily numbers the total downloads is about 1.7 million for the 4.1.12 release from May.

> but I wouldn't be suprised if there are windows users who are still downloading OpenOffice

Definitely. I have been using OpenOffice for years until I found out by accident that LibreOffice exists... It's a shame!

Saw openoffice on a friend of my father's laptop a while back. it had a 4k screen and OO did not support it, causing every pixel to be 4x the size of other applications, it looked really awful. I told him to download libreoffice instead and that openoffice was not being developed anymore. He thanked me and told me that he had been annoyed by how awful it looks.

so yeah, it definitely happens

I bet if I polled friends and family the ones who know the name OpenOffice would outnumber the ones who know LibreOffice 2:1. And it's only that favorable for LibreOffice because I know quite a few other computer nerds.

Thank you for your post. I am one of those users and I'm downloading officeLibre now. Open office has been usable but plagued with random issues.

> I wouldn't be suprised if there are windows users who are still downloading OpenOffice

I'm far away and can't help my parents. They recently replaced their computer and asked a local tech to reinstall their apps. I told them to ask for LibreOffice. The guy put OpenOffice...

The damage from Apache is real and massive.

The mark of a good open source maintainer is the ability to, when your personal interest wanes, find a new maintainer and hand over.

I've done this several times, and certainly development went on much slower than before, but them's the breaks. At least I did the right thing.

For such a big organization, with so many open source projects under its big wing, Apache has failed on this mark of maintainership completely. It failed so badly that at this stage, deploying Apache-maintained software for a customer or employer is downright irresponsible.

Would you organize all your company's internal communications around Google's latest messaging thing? That's insane, you know it'll bite you in the ass in a few years time when it gets cancelled. Apache is no better.

The ASF is an umbrella organisation. For most projects the ASF’s involvement stretches only to procedures and structures for managing the project. The maintainers are mostly not even ASF Members (even those that steer projects). So, not much inference can be drawn from the failure of one project that applies to another.

To be honest I never understood why Apache even started in the first place. By that time the LibreOffice fork was well underway.

"“What just happened here is that Oracle couldn’t resist giving the community the finger one more time,” https://www.linuxinsider.com/story/oracle-and-openoffice-the... (2011)

Sounds like Oracle, with their behaviour with, well, everything all the time.

The whole OOo, mysqL, OEL junket within a span of a few years, made sure I'd never use them, ever, for anything.

These days, I think it's maybe in alignment with why nigerian style scam spam is always so poorly worded, and looks so obvious.

They want the easy marks. They want those not easily fooled, to not reply.

And so this is how I see Oracle, and their horrid, childish, terrible behaviour. The want to weed out anyone who senses the scam, be it licensing hell, or just horrid core/services.

Sadly, big corp and governments are slow to get the message, and I suspect more easily fleeced.

The amount of times I've heard of a CEO or CTO in big corp, buying a product without even talking to devs/IT?

The scam lives on.

Depends entirely on the project. Java (probably because oracle doesn't solely own it) has flourished primarily under their hand.

And how many times did oracle java have regressions for timezone changes?

I hate them for my 2am(1am?) emergency daylight savings calls.

Oracle java flourished, but part of that has been suing everyone under the sun for using it.

Look at android/google...

More scam, more or the same from oracle.

> And how many times did oracle java have regressions for timezone changes?

Did other languages handle it better? Is this a Java problem or is this a "Governments changing timezone information all the time is hard" problem?

> Oracle java flourished, but part of that has been suing everyone under the sun for using it.

> Look at android/google...

AFAIK, android/google is the ONLY ones they've sued. Do you have other examples?

> Did other languages handle it better? Is this a Java problem or is this a "Governments changing timezone information all the time is hard" problem?

Java is peculiarly annoying in that it doesn't use the normal tzdata database, it has its own copy of the database, with its own separate updating procedure. Even when it's the openjdk package on a Linux distribution, you often have a "tzdata" package (used by everything else) and a "tzdata-java" package (used exclusively for the JVM).

So yeah, cross platform is hard and this was their implementation of it? How do other languages handle this cross platform use case? I'm just being pragmatic here and less dogmatic. They chose to manage their own tz records so that the host is could be broken to high hell and theirs would still work. My assumption of this is having to support N different is and arch differences that made this non-trivial when they adopted it. Maybe now there are easier paths to take, so I suppose patches welcome?

Its one of those things where a switch to use the host tz database would be welcome

Everything else uses the system timezone database wrapping whatever api the os provides.

It was before Oracle, but I'm reminded of Sun's lawsuit against Microsoft over their Java implementation.

https://en.wikipedia.org/wiki/Visual_J%2B%2B#Sun's_litigatio...

Right, which is why I think it's a little crazy to talk about how litigious oracle is with Java when there's no major cases I can think of besides the google and microsoft cases.

It's been over 10 years, you'd think there'd be far more examples of bad behavior.

What's the problem with MySQL?

Since it's a database system it competes with Oracle's paid offerings and anyone aware of this conflict of interest has been waiting for them to hobble the product.

Now there is Mysql "enterprise edition" and "community edition" - but paying can open you to a license audit from Oracle. Savvy people just use MariaDB and avoid the whole mess.

I'm responding to a claim of "The whole OOo, mysqL, OEL junket...", and all you've got is to accuse Oracle of what they might do in the future? It's fine to hold a negative opinion of them because of what they might do, but I don't think it's fine to use MySQL as an example of something bad they've done already...because, as far as I can tell, they haven't.

> Savvy people just use MariaDB and avoid the whole mess.

I'll ask again: what mess?

> what mess?

I just described it to you?

A few years back, Oracle created OEL, that is, Oracle Enterprise Linux.

However all they did was to effectively clone Redhat's src rpms/repos, and call it a day.

Later, as their stuff was 100% identical, they just used redhat's rpms for security updates. Essentially, one paid corp contributing not a single thing, and completely and totally leeching off of redhat.

Redhat, whom they wanted to take market share from.

The next step was to aggressively market OEL, and of course claim that it was better. Of course Redhat did a sensible thing, and moved its security patches behind a wall, and required a subscription to get them. Subscriptions were free, but the terms meant Oracle couldn't just copy them. And of course, Oracle couldn't tell people to use Redhat's url/servers for updates on the OS they claim to have "created". Bad image, and all that.

Angry at the idea of having to manage their own security updates, and you know, do actual work, Oracle hid all security patches, and security update info for mysql.

This made it impossible for not just Redhat, but everyone (eg, debian) to backport security fixes.

And there is no equivalency here in Oracle's response. After all, Redhat security updates were created by bug reports about CVEs, and patches in software (eg, a ssh or apache patch), which Oracle could easily obtain and do themselves.

Instead, Oracle stopped communicating specifics about security issues, and providing any patches for security updates, purposefully obscuring security fixes in massive single patches, with loads of other fixes and changes.

This is the reason the entire open source community embraced mariadb, everyone had no choice.

It is also the reason redhat threw money at centos, and essentially bought them. This allowed them to give security updates to centos directly.

All this because Oracle threw a temper tantrum, at not being able to build an entire OS, for free, and leech off of the company they wanted to take market share from.

Oracle is rotten through and through, pure scum.

I wonder if there might be an article somewhere so that I could read more about this.

Still, none of this relates to MySQL. You can create a list of things you aren't happy about Oracle for. MySQL isn't one of them.

Making it very difficult for people to manage deploying mysql, as well as making it hard for security researchers to audit change, and security updates, is a real problem.

But that's oracle for you. Buy something, mismanage it, lose market share like crazy, the only plus is, it is amusing to watch them fumble about I guess.

They basically bought something, then lost 99% of their market share. Essentially every linix distro dropped mysql. It is gone from redhat, gone from debian and all derivatives.

Mariadb, Monty or not, was just a startup when all this happened. They literally handed Monty billions on value.

Pathetic.

Meanwhile, so many changes in MySQL weren't even from Oracle after they bought it. Loads of stuff came from Percona. Massive code for 5.6.

Oracle makes Facebook look like a model citizen.

> Making it very difficult for people to manage deploying mysql...

How so? You can download source "from upstream" https://dev.mysql.com/downloads/mysql/8.0.html?os=src or enable their apt repository from https://dev.mysql.com/doc/mysql-apt-repo-quick-guide/en/. This is pretty much what every upstream does.

> Essentially every linix distro dropped mysql. It is gone from redhat, gone from debian and all derivatives.

False. It's available in Ubuntu in the usual way. "apt install mysql-server" is all you need. Most of Ubuntu builds against MySQL, not MariaDB. MariaDB is also available.

As for the others, being removed from other distributions is hardly Oracle's doing.

> Meanwhile, so many changes in MySQL weren't even from Oracle after they bought it. Loads of stuff came from Percona. Massive code for 5.6.

I don't think this is true. For example, the replication in MySQL 8.0 is unique to MySQL. Many changes fix dubious design decisions from previous releases. Every release seems to be full of old ambigious or poorly engineered stuff being deprecated or removed.

It seems to me that you dislike Oracle because of what they did with other products, and are mistakenly blaming Oracle for mistreating MySQL but can't actually point to any MySQL specifics here. Please don't make your argument circular. If you don't want to touch MySQL because of other things that Oracle have done, then that's your choice. But putting MySQL itself in the list of things that Oracle is doing wrong doesn't stand up to scrutiny.

Oracle acquired MySQL in 2010 when it bought Sun. It has played stupid games in the intervening time.

I do wish they had done the same with Solaris or ZFS. They could have at the stage given it a permissive license and then the community could have kept that or went to GPL and killed the whole CDDL/GPL debate, but now the non-Oracle-assigned changes are also under CDDL so that window has closed.

IBM was shipping their own proprietary office suite based on OpenOffice, Lotus Symphony, and LibreOffice wasn't a good fit for that business model. IBM was the leading charge behind Apache OpenOffice, Oracle would have rather dumpstered it.

I'd be surprised if the walking dead state of Open separate from Libre wasn't caused by in-house tools somewhere that just never got ported to Libre, or that are somehow inherently incompatible with that fork. Symphony is certainly a prime suspect for the original "version some in-house stuff got built on".

The question "do we really want to port yet another minor improvement/fix from Libre to Open instead of porting our own grown mess from Open to Libre?" can be answered with a short-sighted "yes, of course!" any number of times. And Open cannot be formerly declared dead as long as this form of contribution keeps happening.

IBM discontinued Lotus Symphony nearly 10 years ago. OpenOffice continues because the Apache project chairs in place hate the GPL, and even worse, they hate admitting they lost. It's a moral/pride thing rather than a practical or business thing at this point. But IBM was the reason it exists in the first place.

Judging from the simplified fork graph in the article (likely greatly simplified, I admit), I'd expect something depending on Symphony to be far more compatible with current Open than with current Libre. But sure, the licencing difference might very well be a part of that.

> OpenOffice continues because the Apache project chairs in place hate the GPL

LibreOffice isn't even GPL. It's Mozilla Public License (MPL). Still copyleft, but much more weakly so, because only MPL source files are required to stay MPL. You can mix MPL files with even proprietary source files, and you only have to distribute the changes to the original MPL files.

Though I suppose that might conflict with the Apache foundation in general. Don't they require Apache license for all their software?

What "business model"? IBM didn't charge for Symphony, and if you wanted to pay for support it was fantastically inexpensive. Even cheaper than I remembered, https://www.computerworld.com/article/2535199/ reports 20k users for $25k/year.

So what business model was libreoffice disrupting, exactly? Because IBM at the time was mostly driven by NIH syndrome rather than some 4d-chess anti-freedom scheming.

I'm in the business of building a word processor.

Besides OpenOffice, there is generally reduced downloads of both OpenOffice and LibreOffice. This is not surprising.

The world is moving past system-installed, standalone editors, saving them into a file and exchanging them over emails. The convenience that comes with editing, sharing and collaborating right from the browser is picking up speed, and it's a good thing. We have unlimited version history and the social features (comments, reactions and data tables, with ratings and upvotes as column types like in Coda / Notion) is unmatchable with what a traditional word processor can offer.

The world is evolving into a different future of Office apps and Apache's projects don't fit into that ecosystem.

It’s NOT a good thing. I want to own my documents and not have to worry about some cloud provider mining them to sell data about me, or just disappearing one day.

Sure I can see the cases where the collaboration is useful, but there are many, many more cases where it’s not needed and cloud introduces far more risk than reward.

It’s also hard to beat the UX available to native applications that are dedicated and purpose-built and locally processing rather than an interpreted language wrapped inside a web browser constantly communicating with a server somewhere as I type.

I really hate online document editors.

How many documents do you actually own, but still write them in a full-out word processor? Personally, if it's just for me, I use .txt files. Word processors are too much friction for personal notes.

The only cases where I want sophisticated formatting features are for other people's documents, which means it's their business to pick the format. They don't always pick Office 365. Sometimes it's Confluence, or a company-branded tool that I don't even know what it actually is, or Zoho, or G Suite.

Between running some random word processor that someone else picked for me in my browser, and giving it unrestricted access to my home folder (or having to fiddle around to get it running in a container of some sort), I prefer the browser.

I definitely prefer the browser over imposing a monoculture on word processors. Not so much because I think there's much room to innovate in text editing itself (I often enough copy-and-paste between web-based word processors and Sublime Text), but because of the opportunity to innovate in the data model.

no, get your social crap out of my spreadsheets damnit

Few regular folk need to share documents and businesses will use name brand word processors for collaboration rather than some unknown, new variation. You're wasting your time, IMO.

Does Apache ever "declare" anything dead? In my experience things are just allowed to drift quietly into bureaucratic ghost towns. I'm not sure if they've ever formally disbanded a project, but maybe somebody has an example.

It would be helpful if OpenOffice were formally killed though. The name "Libreoffice" is not particularly catchy and I can never remember which is which even though I actually have LibreOffice installed on the computer I'm using right now.

Edit: Oh here's an example: https://river.apache.org/ but it took them several years to come to terms with it, but then again that was quicker than OpenOffice, yes.

Edit again: I think coincidentally (or not) Apache River has a similar history to OpenOffice, in that after buying Sun Microsystems, Oracle "donated" it to Apache, "donated" meaning "Good luck, this isn't our problem anymore." One of those things that looks like goodwill but really isn't...

IIUC, the central Apache organization can’t or at least doesn’t declare projects dead. I’m not sure what would happen if the maintainers literally disappeared, but as long as the project’s PMC stays active and wants to keep going they can.

They do declare projects dead. They even have a place for them https://attic.apache.org/

I use several Apache libraries that are getting pretty old. But so is my code. So I'm glad that they keep things around a long time.

Still, I think it might be good to have more warning for the consumer-facing tools.

Apache moves projects to the "attic" regularly. It's a well known process but requires the project community to be really dead. Which is not the case of OpenOffice just yet. See https://attic.apache.org/

How is "really dead" determined?

The project chairs vote to close the project.

> When should a project move to the Attic?

> Projects whose PMC are unable to muster 3 votes for a release, who have no active committers or are unable to fulfill their reporting duties to the board are all good candidates for the Attic.

PMC = Project Management Committee

Apache could at least put a "use LibreOffice instead" notice at https://www.openoffice.org/. Why not?

Others have mentioned the Attic, but I think the other crucial thing to be aware of here is that to the Apache Software Foundation, it's all about community. When a project wants to graduate from the Incubator and become a top-level project, no one asks what cutting-edge technology and new innovation is available for download, they ask about the health of the community. Are there enough active contributors? Do they come from a diverse set of employers? Is the mailing list traffic stable? Are you actually making releases and getting people to vote on things? There's no rule about how much work has to be in each release. That may, as you say, lead to bureaucratic ghost towns where the technological innovation has stopped, but that's why. That innovation is up to the individual communities. And as long as the community is there, Apache governance will not just barge in and tell you you're done.

I was on the PMC for a project that is now in the Attic. The project had been dead for a long time already. There was a discussion about killing it - only one company still had any interest and even they weren't actively doing anything. Apache wanted to see 3 active PMC members, so I volunteered to be one of the 3 if other people were interested, but then the next time they asked it was clear that no one was actually going to do anything. No releases. No mailing list traffic. No votes. THAT was what killed the project. The project had basically stopped innovating soon after graduating to a top-level project and had been that way for years - but there WAS active maintenance and an active mailing list of users. And Apache is fine with it - it is, as they say, a natural part of the life cycle of a project to go into maintenance mode.

As long as people are responding on the OpenOffice mailing list and releases are happening, Apache can consider that community alive. That they're not keeping up with LibreOffice or making major releases is not really a big concern. If the community really dies, yeah that's what the Attic is for.

That said if they don't plan to change, I'd love to see the community endorse LibreOffice as a successor.

That they're not keeping up with LibreOffice

Who cares about "keeping up with LibreOffice?" The idea that this is a competition is an idea that seems to only be constantly promulgated by the LO camp.

AOO exists and releases on the pace that they can. If that isn't sufficient for some people, those people are welcome to jump to LO. For people who are happy with AOO, so be it. How about a little more "live and let live" here people, instead of the LO camp always trying to bully AOO out of existence just so they can declare final and complete victory? Maybe they'd like to drink the blood of the AOO developers out of their skulls as well just to emphasize the point?

I literally said "X is not really a concern" and you come back with "Who cares about X?" and something about drinking the blood out of skulls? Whoa. Slow down a sec.

I don't see anyone rooting for AOO to die. The only reason I see here for people to care about keeping up with LibreOffice see that AOO got a much better known brand and clearly haven't done as much with it, causing people to try out an open-source office suite and have a subpar experience with FOSS.

I was just using what you said as a "conversation prompt". My post wasn't really a reply to you per-se. I'm talking about something more abstract, and about the people out there who do treat "keeping up with LibreOffice" as some kind of big deal. Apologies for the confusion.

I see - my bad. Ignore the first paragraph, then. Second half still applies I think :D

The problem is that the users are going to AOO for historical reason (brand recognition, basically), and getting a drastically outdated product.

While the OP says the last major release was in 2014, I just checked and the last patch release was May 2022. So it isn't dead-dead, or even brain-dead. I've seen worse.

So, yeah, you make a perfectly valid point.

Here's what I suggested as a goal for OpenOffice, in September 2016 which is now six years ago, as a response to Apache officer Jim Jagielski ("jimjag") who thought this was going well:

[[[ I urge you, or the Apache board generally, to look at the facts, and not get yourself into the mess of trusting AOO project members' gut instinct that everything is fine or that the board's oversight is the problem and you should just stop looking. For sure, it is possible that they'll actually manage to turn this around. But there are already AOO people convinced they've done it, and there were six months ago, when in fact as you know they were in deep trouble already. Their feelings about this aren't going to be objective. If in say, three months, there's no measurable evidence that AOO is back on course then regardless of what is said by the handful of AOO people, retirement is the right choice.

For example, shipping AOO 4.2 in 10 weeks at ApacheConEU. That's not crazy. Libreoffice goes from feature freeze to release in 10 weeks. A healthy AOO development community should be able to do it, or come so close as to leave no-one in any doubt. ]]]

https://lwn.net/Articles/699680/

Again, I wrote that in September 2016. Did Apache in fact ship OpenOffice 4.2 at that year's ApacheConEU? No they did not. OK, so when did they finally ship the "imminent" OpenOffice 4.2? Never.

Here's what the most recent quarterly summary from the Apache OpenOffice project to Apache's board said about version 4.2, "An alpha and/or beta release is planned for the next quarter". That sounds optimistic doesn't it? Well, sadly that's also what it said in January 2020. Huh. Who writes this stuff? Funny, it's Jim Jagielski, the same guy I was addressing six years ago.

Try to avoid caring whether people are incompetent or liars, you will find it difficult to tell which is which, so if you ensure you can treat both the same you needn't distinguish. In this case I needn't worry about whether Jim was too stupid to know what the problem was in 2016 or just didn't care.

Is Jim getting paid to work on AOO?

He works at salesforce atm

Apache has a few good projects and so much abandonware it’s very difficult to know which to avoid entirely.

At this point I think Apache has about 15 different data streaming/queueing/event processing systems.

At a minimum, and they all have so many bugs :(

I often feel like projects that are handed over to Apache are basically put in the equivalent of an aircraft boneyard; stashed for archival purposes, but it will no longer receive much development effort.

Seems to be the pattern. Whenever something gets the "Apache" prefix I get nervous that it's abandoned.

Giving Apache ownership of something is always a signal that whoever was maintaining it before, doesn't want to maintain it anymore. Whether or not that thing dies depends entirely on who ends up running the project post transition to apache.

Funnily, there have been a few cases of projects that have seen a breath of new life by moving over to apache. Netbeans and Groovy seem to be standouts there.

> Netbeans and Groovy

I used that ten years ago

I though they were dead by now

Around the time they transitioned to Apache, they were pretty close to dead. After the transition they've both notably picked up dev work.

Kafka and CouchDB are both actively developed

Cassandra and most of the Hadoop zoo too, but that's mostly because of a few vendors. (Eg. DataStax for Cassandra, and Cloudera & Hortonworks for Hadoop. Plus probably a bunch of smaller ones.)

Companies handing their failed projects over definitely seems to be a pattern.

On the other hand the ASF has a few very active and healthy projects under its umbrella, sometimes being among if not the go to solution inside a specific problem space.

Come again? Lucene, Solr, Kafka, Zookeeper, httpd, CouchDB, .. and that's just the top of my head.

It's a misconception of what Apache is. It's legal umbrella for open source projects, not an organization with a goal of actually developing projects.

Good point, and fair.

Apache presumably has criteria for accepting an open source project? Although all I can find is this, marked "draft": https://www.apache.org/dev/project-requirements

But I'd suggest they need some, and also minimum requirements for remaining an Apache project, which if not met, the project will be either removed, or perhaps just clearly marked archived/unmaintained/historical on apache pages. (a status from which perhaps there is a way to escape with renewed attention as well).

I think Apache Attic is exactly this.

https://attic.apache.org/

In the case of OpenOffice, they've repeatedly gamed efforts to close them down by doing bad-faith "releases" to pretend that they've got activity.

ASF governance is weak, and that's why it might as well be the Abandonware Software Foundation.

That's a consequence of the ASF largely serving as a way to dump software that isn't commercially viable (or wasn't to begin with). The idea being that a community can form around it, consisting of former users and developers. Taking software to the ASF is accompanied by a grant to the ASF, and I think it's a decent way to stop developing a product, or is at least miles better than a torched earth approach and stop sales, support, development immediately, firing developers, etc. Not everything will live on ofc, but at least one can try.

Is ASF the CA of Free Software?

Softbank

Do you mean certificate authority or California? I can see an argument for both.

Probably Computer Associates. Speaking of soulcrushing software graveyards, there's also OpenText (though it is less well known).

> Probably Computer Associates. Speaking of soulcrushing software graveyards,

The Google of Free Software then.

“OpenText”

(gets PTSD shivers thinking about LiveLink)

TIL.

Oh, and that was Tim Bray.

Bingo.

CA is where software goes to die.

Yeah you're so right, everybody stop using this Kafka abandonware! Nothing to see here...

They have Google Wave -> Apache Wave

The practical answer is that you go and look at a project's release history at its primary repo. I don't know of a good way to browse all apache projects by livelihood though. The closest thing I know of to keep track of Apache projects overall is this:

https://lists.apache.org/list.html?announce@apache.org

But just eyeballing that really quick I can see it's not authoritative, because a new version of Apache Spark just came out and isn't announced there.

https://spark.apache.org/news/index.html

There is also this, but it's very cluttered with submodules, but it does appear more authoritative.

https://projects.apache.org/releases.html

I've learned over the last decade or so just to avoid Apache software. It seems like the vast majority of their software is poorly written, enterprise abandonware. They are a joke of their former self. Kind of sad really.

What are those some of the good Apache projects?

Practically never used anything but httpd (which I no longer use) and everything else looked like bloated software being Java taking huge resource and usually has a better competitor.

I occasionally see Kafka being used as parts of software stack but not sure if that is the best choice with huge resource use.

Their Java libraries and utils are good: Maven, Utils, Commons, CommonsIO

I see a lot of blah blah blah below, but none of it gets to the crux of the matter, and that is LibreOffice is klunky.

I used to have it installed, because I had installed it for my father-in-law and I was his tech support (he passed a couple years ago).

The API in Writer is a mess. Okay? I am sure I'll get some people sqawking for be saying that out loud, but I also know a bunch of you will be nodding in agreement.

And don't get me started on Calc.

Lets get another thing clear, I don't love Microsoft Word. I don't. But then, my original experience writing documentation was in nroff/troff using Writer's Workbench. Then later, TeX. But the truth is, through work, or school, a ton of people have become accustomed to be productive in Word.

And Excel is simply Superior to Calc in every way.

And yes, Microsoft is up to their old tricks, the bastards. We all know it. I mean, WSL2 was pretty good in Windows 10, but OMG WSL2 in Windows 11 competely rules. Native Linux USB support. Wow. That is some wicked bundling there. And then, if you have a family, Office 365, just for the fact that you can hand it out to a total of five other people, and you each get 1TB of cloud storage. For $99 per year. Again, wicked sneaky bundling. Six instances of 1TB cloud for $99? We only use five here, so $20 each, per year, for 1TB of cloud. Oh yeah, you get to use Office 365 if you want, as well.

So year, Microsoft pulls under handed market bundling.

But that doesn't excuse the fact that LibreOffice is Klunky. Now that I no longer have to support it, I don't miss it at all.

(shrug) I still use OpenOffice 4.1.5 day to day on windows 10, with no issues...didn't know about any controversy.

There is no controversy: OpenOffice is crapware, use Libreoffice.

The only problem is that Apache still distributes what they are fully aware is an abandoned husk, misleading people like yourself.

This comment does not explain why one should use Libreoffice if Openoffice is working just fine for them.

security, probably? using abandonware to open extremely complex file formats like microsoft/opendocument ones is kind of a bad idea.

Yes. Do you absolutely trust everybody who sends you a doc or spreadsheet? Do you trust that much everyone who sends them one, that they pass along to you?

It is better not to need to think about that, much.

This is kind of like asking why one shouldn't run Microsoft Office 2003 anymore.

Does it still work? Sure... is it secure and has feature parity with modern offerings? Definitely not.

Perhaps that's fine - but since LibreOffice is also free and is actively developed (by many of the original OpenOffice team too), I don't see a compelling reason to not make the switch.

Wow, didn't realise StarOffice started in 1985.

Online office apps have an edge WRT collaboration. It may well be that the era of the standalone office suite is reaching its end.

What would it take to make LibreOffice a web-centric app companies could host themselves? Maybe Apache's resources could be directed that way.

I think this exists in the form of Collabora Office

https://www.collaboraoffice.com/collabora-online/

It is integrated into NextCloud I believe for self-hosting

I am not sure that online collaboration is such a killer feature.

Most of the time I would take a good offline office app over the application that requires an internet connection, stores my data in the cloud and forces its design choices, from the UI to the security, down my throat "because of the way cloud apps work". My 2c.

This is probably a plurality opinion on Hacker News, which means it in no way represents the opinion of the average user or business professional.

https://books.google.com/ngrams/graph?content=microsoft+word...

Note that 2010 is the year that MS first started adding cloud-based collaborative tools to MS word, reducing the downward trend.

LibreOffice does have some form of online version: https://www.libreoffice.org/download/libreoffice-online/

To add onto this, I have been using Libreoffice Online or Collabora Online for the last 4 years or so. I used to compile LOOL myself but eventually I just switched over to using the Collabora Online Development Edition that Collabora releases. I am using it integrated with Nextcloud.

I enjoy using it and it has been a life savior for the times I need to work on a document with someone else since I can just send them a share link for it. The fact that it uses leaflet which is meant for maps does mean that the latency to the server matters quite a lot to experience but I don't mind that much, personally, with my not great 120ms RTT to my server.

Hey - thanks for using Collabora Online (COOL) - we're doing a chunk of work to accelerate interactive editing, and lots of that starts to land in 22.05 - time-based document change compression (deltas), improved JS, reduced latency of event handling (particularly in the browser where handling websockets related to rendering & the DOM is a bit too 'exciting') - with lots of wins recently. Hopefully you can enjoy them in our next release coming soon.

Honestly I hate online office tools. They encourage a collaboration model that just doesn't work for me _at all_ - and yet is, by now, _the standard_.

I'd much rather send documents back and forth, with comments I integrate with the material. If I _need_ to do something live, plain text is actually not just perfectly sufficient - it's either _exactly_ what I want, or what I want is a white board (and therefore please let's just use self-hosted Excalidraw - but remember to sponsor it if you're commercial!).

But outside of very rare occasions, let's just set up a plain old mailing list, write proposals in basic text, and then once we have something complete, let one person integrate the whole thing into a single pretty document with like LaTeX or Markdown or something. Or even Word, I don't really care.

And yes, that's hostile to brainstorming. Good. I too am hostile to brainstorming.

Online apps have an edge for collaboration.

They have a downside for privacy and security.

What I'd really like is on office suite that lives on my own device (e.g. my router), where I own my data. That's not the way the world went, but c'est la vie.

Synology has a full Google Drive replacement with apps built-in.

I think the fact that I'm The first person to mention it here, and the fact you didn't find it yourself, is revealing about the true level of demand for such a product.

I think it's indicative of marketing efforts. Their wikipedia page doesn't mention it:

https://en.wikipedia.org/wiki/Synology

The products menu on their web page doesn't mention it.

https://www.synology.com/en-us

IF you go to Solutions -> Data Management -> Office, you find it. One more link, and you're there:

https://www.synology.com/en-us/dsm/feature/office

Of course, there's no way to try it or to buy it from there....

I see random devices for sale for under $200, but I'm not sure if they include it. If I bought it, I'm not sure how well it works.

We have no idea of the level of demand for a product that well-hidden. The fact I just spent 5-10 minutes without figuring trying to figure this out without success means we have no idea.

As a footnote, I do have a random machine running a random online office suite. It works, but not well enough to be generally usable.

> They have a downside for privacy and security.

What if you could host it on your laptop or a server you pay for?

When I suggested something a company could host I imagined it replacing Office 365 or Google Docs (a concern governments may have about having American companies host their data). Being able to self host and allow collaboration without relying on a third party application is important.

There's definitely a significant niche for offline word processor, spreadsheet, drawing diagrams and such. I do not want to go online, register, have my data "in the cloud", fume when it goes down because of AWS of buggy update, lose it when the company decides to fold, etc. I want to fire up an application on my local, edit it there, and save it on my local. If I need collaboration, well, there's plenty of tools for that.

But I'm old and my 14yo son never saw even MS Word (leave along LibreOffice) until I showed it to him, they do everything in Google Docs and that's all they know. So maybe I'm part of a dying breed :-)

Because of what’s been highlighted in this article I found it super difficult to figure out a good alternative to MS Office, so agree 100% with the article. Now a happy LibreOffice user.

It's not dead. I use it. Not that I dev for it though.

Why are you using it instead of LibreOffice? Genuinely curious.

He may not know why he should be using it instead of OpenOffice.

I’m also curious. Also compared to OnlyOffice?

OnlyOffice is far from being as comprehensive as LibreOffice. At least when I tried the spreadsheets, I ran into limitations even though I use spreadsheets very rarely.

Good to know, thanks.

They (he, she, undetermined) might not even be aware of the other office suites. The sibling commenter lists OnlyOffice, first time I hear about that.

1) No Windows release afaik. But please correct me if I am wrong -- I don't see a Win download on their page from a quick look on my phone.

2) I did use it on Ubuntu sometime in college (around 2016 for class) and I just didn't like the feel of Libre. OpenOffice still had a lot of feature parity with Excel without buying Excel, so I got that instead.

They've had Windows builds for as long I can remember. https://www.libreoffice.org/donate/dl/win-x86_64/7.3.4/en-US...

There should be a windows release, its on the downloads page.

>1) No Windows release afaik. But please correct me if I am wrong -- I don't see a Win download on their page from a quick look on my phone.

It's been available for Windows for years. The first option on the drop-down menus says "Windows (64-bit)"[0]. 90% of their 10 million worldwide users in 2011 were on Windows machines[1].

[0]https://www.libreoffice.org/download/download/

[1]https://en.wikipedia.org/wiki/LibreOffice#Users_and_deployme...

> I don't see a Win download on their page

On the download page, just above the "Download" button, there is a dropdown to select the installer. There should be "Windows (32-bit)" and "Windows (64-bit)" there along with the Linux and MacOS installers.

https://www.libreoffice.org/download/download/

Others have already noted that the windows release is easily findable, so instead I will ask: what features are missing from LibreOffice that OpenOffice has? The former is a fork of the latter, and LibreOffice has undergone extensive further development where OpenOffice has stagnated.

After checking this thread, I downloaded Libre on Windows and played around with it a little, to refresh my memory. My main use case is just working with some personal spreadsheets in Calc, no complicated work stuff.

I wouldn't say features are missing, if anything it sounds like Libre is way more fleshed out than OpenOffice. That said, I still prefer the (relatively) legacy UI in Open Calc than the slightly more modern UI in Libre Calc, just a personal preference. It doesn't seem that much more advanced, and after all these years an ODS spreadsheet still looks like a spreadsheet. Other people noted the security posture of Libre though, which is a great point.

Just wanted to chime in that your reasoning here is 100% valid.

I'd also be curious about something: suppose you used LibreOffice for a week or so for working on your personal spreadsheets. What percentage of your work time would you waste either searching for the new location for the old command, or learning the new command that maps to the old one?

I'd love to read a case study where someone has documented this with a UI change in FOSS.

For me personally? 0%. I just do simple copy/paste, scatter plot, sort asc/desc... not a heavy user of hotkeys. I had some trouble finding the Chart button at first since the icon's different, but that was the only issue.

Not OP, but I know sometimes I forget I want LibreOffice. Many years ago I worked for a computer store where I used to refurbish computers everyday with OpenOffice so open OpenOffice is still the brand that first comes to mind for me. I never get very far though because as soon as I see "Apache OpenOffice" I realise my mistake and search for LibreOffice.

But I'd assume lot of OpenOffice users today are just legacy users who use it just because they're familiar with the brand and that's what they've always used.

If you type "apt install openoffice", don't most distros install libreoffice instead?

(not the poster) one of my problems with LibreOffice is that it sucks when you are not using English as your main language - little support for spell checkers, etc. At some stage open office used to be better in that respect!

Not the gp (and not IT/Tech based) but I have OpenOffice available because...

1) editing documents I wrote some time ago. There was a time some months ago when LO would warn you when opening an older oOo doc about changes in the format. I saw this with drawings and Impress files. (I need to see if this still happens with latest LO).

2) one of my employers has oOo on their standard desktop image alongside MS Word and all. So I want the same as them for their documents

3) I know the menus and the little UI glitches well in oOo having used it since StarOffice days. LO is (entirely reasonably) refactoring stuff and therefore changing the ways some features work (drawing tools I recollect)

4) gives the impression of running snappier on an ancient laptop that I use. LO does seem to need a full DE (e.g. Gnome/KDE/xfce)

...if oOo life support was switched off, I'd shrug, move all the documents over to LO, and move on I suppose.

It is already switched off, long since. Except downloading.

I'll nag employer1 again over the summer.

Can anyone point to an actual security breach that is traceable to OpenOffice? That would focus minds perhaps.

> LO does seem to need a full DE (e.g. Gnome/KDE/xfce)

LibreOffice seems to do just fine for me here without a DE. I wonder if there's something I'm missing that I don't realize?

If you don't mind me asking: why? What does it have going for it that LibreOffice doesn't?

openoffice came out years before libreoffice, so if you had learned to work with it at the time and you still find it sufficient, then why switch? I doubt libreoffice is 1:1 identical in terms of UI/hotkeys/etc

Because it's basically an outdated version of the same software. Do you also keep on using Word 2003?

A fair number of people still use Word 97. It does what they need and still works on modern systems. Why switch?

> A fair number of people still use Word 97. It does what they need and still works on modern systems. Why switch?

Perhaps so you can open all of those .docx files that people send to you?

There are people who don't need to do that. Their use case may be writing the monthly newsletter for the local rabbit breeders club, or writing that novel they've been working on for decades now (and totally are going to finish any day now! I can relate, because I've got two or three of those saved somewhere).

To these people, there are no benefits in upgrading.

(Also, I've spent a lot of time teaching my parents not to open office documents attached to emails, as that has become a popular way for people to spread malware.)

> Do you also keep on using Word 2003?

Playing devil's advocate: if Word 97 (or something even older like wordperfect) does all you need, why should you upgrade?

Things are different, of course, when you have to interoperate with other people (in the word processor case, mainly by reading files written by other people); not only do security fixes become more relevant, but also newer versions of the software have greater compatibility (in the word processor case, not only there was a lot of work on libreoffice towards better compatibility with Microsoft document formats, but also AFAIK newer libreoffice versions write by default using a newer version of the OpenDocument formats, so if you have an outdated word processor, you might have difficulty reading files written by other people using these newer versions).

no, I use Word/Excel 2007. and Photoshop CS2. in a Windows 7 VM without networking

they're the ones I've learned to work with, and since I don't use them professionally, they're still sufficient for my needs.

I don't have to re-learn the UI with every other release, and they all run fine with 1 core and 2 GB of RAM allocated, on any machine

> Photoshop CS2

Sounds great actually. Unless I've missed something new, IIRC CS5 is the peak and still unparalleled. Somehow Adobe ruined it.

Depends; has LibreOffice gained any features since the fork? I've had to use it on and off at my previous job and the experience was pretty poor, it ran slowly, dated UI, etc.

The fork happened in 2011, 11 years ago, so there are 11 years of refinement and new features. OpenOffice is in maintenance mode so there are no new features, only bug fixes.

One of the biggest features of LibreOffice over OpenOffice is compatibility with Microsoft Office files.

OpenOffice can open Microsoft Office files but doesn't support as many features as LibreOffice does.

OpenOffice can't save in the newer Microsoft Office formats (docx, xlsx, etc) but LibreOffice can.

Also, LibreOffice now has a ribbon-like interface. To enable it, click View > User Interface and choose Tabbed.

Hi! I'm making this comment via a work computer that has Office 2010 on it. It's not the same thing, sure, but it's a great example of how people are definitely still using old office-related software.

Funny story. Georgia RR Martin writes on an ancient word processor.

https://www.cnet.com/culture/george-r-r-martin-writes-with-a...

LibreOffice was forked from OpenOffice after Oracle bought it. The community switched over to LibreOffice, so it's not like OpenOffice has a longer history than LibreOffice. They have the same history for the first years before the fork, and OpenOffice essentially died as a community project after the fork.

> LibreOffice was forked from OpenOffice after Oracle bought it.

It depends on your point of view. IIRC, LibreOffice is actually a fork/continuation of the go-oo project (https://en.wikipedia.org/w/index.php?title=Go-oo&oldid=10809...), which is older and already existed before Oracle bought Sun.

What happened later is that LibreOffice was rebased on top of the re-licensed Apache OpenOffice, to take advantage of the more compatible licensing. But it had been forked (as go-oo) from OpenOffice much earlier.

You might want to reconsider that, I expect that Apache haven't kept up with fixing the security issues in the OpenOffice codebase.

At one point, an exploitable security issue was unpatched for about 6 months. There was a small note on the site of how to work around it, but it wasn't very noticeable. And every download of it still had the unpatched issue in it until they finally had the resources to release an update. The main purpose of the update was to patch the issue out of the box.

Trademarks. Contracts. Egos. I was there when it all happened. When Oracle refused to transfer the trademarks to the Document Foundation. When loud voices on many sides shouted at each other. Leading to the current situation. My current position: Who cares. ODF and Libre Office's community managed to move forward. Let the ASF have their Open Office.

If trends are telling, as the author seems to think, they tell of Libre's decline, too.

One thing I noticed is that the web based MS Office products are significantly snappier and less quirky than their desktop counterparts. Maybe that is part of it?

Unrelated: I also noticed that Visio, which I previously found quite pleasant to use, actually is nigh unusable (Desktop - no web version exists). Was it always this bad and have my standards simply gone up, or did it deteriorate?

I used to be a big fan of desktop Microsoft office, primarily because of stability, reliability, and ux innovation.

They've clearly stopped investing in quality, every time I go back it is buggier and less reliable than before.

I don't use Visio, but even Word is becoming a nightmare.

This is what scares me: a FOSS project maintained by a corporation. I tend to install FOSS because imo they are more "future-proof", but some of them are developed by companies (e.g., Fedora Linux) and that makes me wonder if they're truly future-proof.

Nothing is future-proof.

Yep. The only constant is change!

The "future-proof" is that if the corp goes belly up or insane, you can in theory fork the project and continue it, which if there's enough of a user base will happen.

But if there's enough of a user base the company probably won't abandon it.

Things like Fedora are potentially at a bit more risk, because if IBM decides to stop developing it for whatever reason most users will probably just switch to an entire different distro (or a related one) than bother continuing the existing.

good readers - I am here to tell you, from experience with my own two eyes, popular important and public software projects absolutely do get closed by private companies, for so many reasons. None of the reasons are "your benefit"

> if the corp goes belly up or insane, you can in theory fork the project

Yes: for a small + useful + easy to maintain project.

No: for bloated or difficult to build projects, which makes most of popular browsers, corp-driven programming languages, devops tools, office suites, corp-driven IDEs, Linux distros...

Not to mention non-copyleft licenses that allow proprietization and discourage communties.

> I tend to install FOSS because imo they are more "future-proof", but some of them are developed by companies (e.g., Fedora Linux) and that makes me wonder if they're truly future-proof.

The story of CentOS should be telling that, no, many pieces of software that are backed by a company will not be future-proof and will probably experience certain changes as a consequence of that, be it being transformed to better fit corporate goals (CentOS Stream), or being retired eventually so the company may focus on something else (Atom), or will just be left to slowly rot over time as happens with most code (OpenOffice).

Then again, it's not like open source projects are that future proof or safe from "drama" either - for example, the Lubuntu project has 2 homepages for no reason: the official one at https://lubuntu.me/ and some other one that serves old versions and is not trusted by my ad blocking solution https://lubuntu.net/

There are also cases, when open source projects experience fragmentation like happened with Gogs https://gogs.io/ and Gitea https://gitea.io/en-us/ and sometimes there are cases where particular individuals simply cannot work together and as a consequence pretty much the same happens, as was the case with Swoole and Open Swoole: https://github.com/swoole/swoole-src/issues/4434

Treat most pieces of software that you use as if they might not be there in a year. Or, alternatively, as if you might need to take over the maintenance of whatever you're using, if you can.

> Atom

Atom should be manageable enough in size that if people really wanted to keep it, they could fork it. Maybe they have.

Correct. If you want reliable and user-centric FOSS the best bet is community-driven.

The second best bet is from a consortium of many companies with a track record of not dropping projects.

Unfortunately most developers still jump on hyped-up corporate-controlled bandwagons.

And apache has sunsetted a LOT of projects to it's "attic".

It even seems it usually is pretty aggressive in doing so.

Maybe there isn't even a person in charge of the project to make the decision anymore, or it's an Oracle employee whose job it is to NOT sunset it.

Why did you even post this? The last release was in May of 2022, not 2017!

https://openoffice.apache.org/downloads.html

Because the Apache Foundation is a graveyard for open source projects.

Because of all the organic internet traffic it brings them. Hey, stay awhile, maybe use one of their ML/AI projects and donate some money too!

I imagine that part of this is due to the nature in which it was donated by Oracle to Apache. Legally obligating it’s continued distribution

Probably Oracle didn't want to cooperate. It's irrelevant though? Anyone who cares works with LibreOffice anyway.

My issue with OpenOffice/LibreOffice is, that it's stuck in the past, which makes me prefer using Google Sheets.

Unless you're in a closed circle who all use similar version of LibreOffice, it's better to use Google's offering or just pay $10/mo for MS' Office and also get 1TB storage, which sounds like a steal and have a less headache with file compatibilities.

I've always assumed when projects go under the Apache.org umbrella, that is where they go to die.

Should I not be using Kafka? Or CouchDB? I guess we shouldn't using log4j at this point though...

I mean yes of course use what works for now, but I think eventually those will die off too. As with most things, Apache projects tend to be bloated Java apps that end up being phased out by more trendy and more performant apps built in Go/Rust.

That does seem to depend on why/how it ends up at Apache. Something like OpenOffice, or the myriad of Java projects, which have just been throw in under the umbrella is more or less abandoned. There seems to be some misconception that Apache has a number of developers sitting ready, or a ton of money to hire developers to take on any random project.

Projects under Apache really only works when the developers are involved and follow the project. At this point I'm a little unsure what The Apache Foundation provides, beyond hosting and some legal support.

The "LibreOffice" name doesn't help. Not that many people would be able to spell that on first hearing.

Libber-what? Lib Re-Office?

Someone should fork LibreOffice every time there's a new release, and re-package it with a better name.

Which better name?

OpenSuite

Freedom Office

Liberation Office

Office Commons

Office Publica

Ensemble

TogetherWorks

> but that doesn't mean that they're unfounded

Yes, yes it does. The very definition of unfounded is having no foundation or basis in facts. That's some "stolen election"-level logic.

But the foundation is there: Microsoft has multiple decade track record of shady shit.

There is no evidence, though. :(

An unwillingness to update a webpage as well as refusing to do bare maintenance for years is astounding. It could have been an unintentional oversight, until the 2020 letter by LibreOffice. (Yes that was not the first to comment). As far as I know, there hasn't been a change - not even an acknowledgement of the letter.

It's been long enough; this is not unintentional and the lack of any response for so long has gone beyond what can be excused by a slow-moving bureaucracy. Moreover, updating the site to acknowledge that the project isn't updated is trivial, truthful, and needful for its users. Therefore that is a more than reasonable expectation for an app with its recognition - even if there were no other competing apps. But nope, nothing.

What does Microsoft have to do with this?

I was addressing your analogy to "stolen election".

At any rate, I'm just pointing out that (1) any decent project would address this situation once they were aware of it, (2) Apache is aware of it, and (3) they had way, way more than enough time to at least acknowledge the issue. Neither ignorance nor bureaucratic slowness can excuse their complete lack of taking any action to protect their users. If you exclude ignorance and slowness as reasons, this is intentional.

I don't know why, but the fact that something is actually wrong on a significant scale and is intentional in no way matches "stolen election logic".

We detached this flamewar subthread from https://news.ycombinator.com/item?id=31840457 and marked it off topic.

Is it me or the second picture or whatever that is doesn’t load?

I get a warning about incorrectly embedded external content, which includes a link to the following KB article from Mozi:

https://support.mozilla.org/en-US/kb/xframe-neterror-page

> Its last major release was version 4.1, from 2014!

Oh we are playing major version minor version!

Let me try: Libreoffice is only at version 7.3 ! That makes it hundreds of versions behind office365! Proof once again that free software has no business competing with software build by professional commercial developers. Now excuse me while I try to reboot my futuristic Windows 2000. /s

Versions may as well be marketing names for most software. If anyone even mentions them I might as well assume a minor case of serious brain damage. Java just flat out dropped the leading 1, Firefox is in the low hundreds, some software just goes year/moth.

As Apache office still releases updates I find the claim that it is dead highly misleading, but that is presumably the intention if the post.

The "from 2014" is the important bit they're emphasizing, not the version number. LibreOffice 7.3 was released this year.

The last release is from 2022 not 2014.

According to Wikipedia (https://en.wikipedia.org/w/index.php?title=Apache_OpenOffice...), the latest patch release was 4.1.11 from 2021, and the latest minor release was 4.1 from 2014. That is, since 2014, they are just releasing small bugfixes. Compare with LibreOffice (https://en.wikipedia.org/w/index.php?title=LibreOffice&oldid...) which had the latest minor release in 2022, and the latest major release in 2020.

Minor versions versus major versions do matter.

Here are the release notes for the 2022 release of OpenOffice:

https://cwiki.apache.org/confluence/plugins/servlet/mobile?c...

This is pretty clearly a minor version update. There are only four entries under their improvements / enhancements section, and all but one would be more accurately described as a bug fix. They even identify this as a "maintenance release".

Here are the release notes for the 2022 release of LibreOffice:

https://wiki.documentfoundation.org/ReleaseNotes/7.3

Aside from the fact that this is much better documented, you can also at a glance see that this is, in fact, a major version update. LibreOffice isn't padding their versions, they actually are adding new features while OpenOffice fixes a few bugs.

> Oh we are playing major version minor version!

I also had some doubts, especially after a quick look at Apache Open Office[1].

Then I checked the announcement of 4.1.{9,10,11,12}. They are all maintenance or security updates.

Compared to the Release Notes of LibreOffice[2], there are improvements with each release.

1: https://www.openoffice.org/ 2: https://wiki.documentfoundation.org/ReleaseNotes/7.4

Reply on news.ycombinator.com