LibreOffice Online dragged out of the attic, dusted off for another go The Document Foundation (TDF) has pulled LibreOffice Online out of its "attic" – its term for retired projects – and is resuming development. LibreOffice Online (LOOL) is the cloud-based version of LibreOffice. You could be forgiven if you haven't heard of it. Its development has been largely dormant since 2020. That will soon …
Well that's a deal breaker right there. As long as Libre Office continues to use menus as the great $deity intended, that's where I'll stay.
Funny, four thousand years or so ago the Mesopotamians invented writing, and now we've advanced back to pictograms (which, in spite of all their apologists, are firmly based on shared cultural norms) and indeed, bombing the descendants of those pesky Mesopotamians. That'll teach 'em.
I don't mind people wanting either. I grew up without one, others grew up with one.
What drives me insane is WE COULD JUST HAVE THE OPTION. Like we did, in fact, used to have!
Rather than fixing something solid and alienating 50% of the userbase... put in an option.
Or better yet go real old-school and program your software to be UI dependent so people can put a UI on top (like, say, a web-first interface!), or they can decide to "theme" it like Office 2000, or run it like Wordstar if that's what they want to do.
Every time, every project, they forget that the ultimate aim of the software is to be USED BY PEOPLE who will naturally have differing tastes.
I moved from Windows when they started telling me where my taskbar had to go, where my icons had to align to, that I can't use folders in the Start Menu but must "search" for everything instead, etc.
Not the functionality or the performance... the UI and UX. And given that programmers appear to be AWFUL at implementing UI/UX, make it modular and let people do what they damn well want with their own computers!
Still don't understand why even Microsoft Word doesn't just have a "Office 97 theme", for example.
> What drives me insane is WE COULD JUST HAVE THE OPTION. Like we did, in fact, used to have!
As my and others' comments bracketing yours say: you _do_ have the option in LO.
In Windows Office you don't. But the Mac versions of the same suite have native macOS menus _as well as_ the ribbon. You can turn off the ribbon. (Although after the first version or two, this does not persist. When you close and reopen the app, bang, back pops the wretched ribbon.)
Since I normally turn off all toolbars anyway, as I'm a keyboard pilot, I am happy driving it with the ribbon off.
The point? Microsoft did have that option. The codebase supports both UI methods.
Same as if you run unmodified Google Chrome for Linux on Ubuntu Unity, it uses the global menu bar. The same binary, menu-less when native, actually is exposing all the API calls to have a menu, but it's invisible, hidden. Run it on a desktop with a global menu bar, bingo, suddenly it has menus.
MS merely wants to force you to use ribbons. It's doubtless patented and protected. Every OS has menus and toolbars now, so the kings of proprietary apps want to compel everyone to use their copyrighted patented trademarked UI. Which is also why most Windows built-in apps gradually sprouted ribbons, making Explorer as unusable as Office.
You are thinking it from an individual user point of view. Many an enterpise/education customer has more than one people using the same desktop (think, operators in shifts). Or machines get reasigned. Or, internal traning and support materials are developed with nice screenshoots to acompany them.
For those customers (who are the customers MS is interested in) is better to standarize in one "true" UI
It also helps that microsoft has to spend less money on development and ongoing support (both customer support, as well as patching and sutff) if there is only one non-customizable interface
That also has the benefit all the adjustments and utilities you need to comply with the DDA should also work for years.
My biggest headache is accommodating dyslexic users across all the various application access methods, then MS changes everything and you’re back to square one.
Perhaps someone should tell MS that ecosystems need a reasonably stable environment to develop and flourish, currently MS seem to be doing everything to destroy the Windows ecosystem; Something the Open Source community should also be aware of and leverage to their and our advantage…
We told them for decades, they don't care.
This is MY COMPUTER. It's my literal desktop. It's where I put things to do work.
And they have never cared about that, when they could splat a weather widget or a sidebar or move stuff around for no reason, or hide most of the older settings so they're really quite difficult to access but still have necessary settings that are available NOWHERE ELSE IN THE OS.
They've never cared about that. It's why I'm always amazed that corporations are so sold into their products. The amount of destroyed workflow hours because of just Microsoft junk that people aren't even using must be into the trillions of man-hours by now.
> As long as Libre Office continues to use menus as the great $deity intended, that's where I'll stay.
This is my own personal attitude, but it is important to note that it _is_ switchable. Normal local native app LO can switch from menubar + 1 fixed toolbar, to menubar + floating customisable toolbars, to a ribbon-like thing...
And although it doesn't work very well in that orientation, the LO ribbon can be placed vertically, which beats the MS implementation.
I have not tested COOL/CODE because I don't run a local webserver and don't want to. (I tested a remote hosted instance but it was about 4Y ago. TBH I do not remember if it has a switchable UI or not.) I would expect it does because it's LO but headless, rendered to the browser.
However Google Apps shows that a menubar/toolbar UI doesn't work great in a browser window. Using the keyboard, you get the browser's menus, not the apps. Using a pointing device, you have to aim twice as carefully because you have 2 menu bars and 2 toolbars (one set real, the other set web content) and you don't want to hit the wrong set.
It is a kludge and not an elegant one.
However, this does not apply to Electron-style apps where the app programmer defines which elements of the browser chrome (the renderer's own UI) appear. They can turn it all off and then the app _looks_ much more like a native one. On some OSes/desktops, anyway.
It does, though it still could use quite a bit of work. Hopefully that's become a priority now that concerns over AI training and data sovereignty are at the front of the industry's list.
Setting up an install for another test drive would make for a really cool article if you haven't used it in many years. AI and data sovereignty are huge concerns now in a way they weren't four years ago.
Took me less than an afternoon to set up, most of which was it being temperamental because we have IPv6 off (this wasn't documented well, and it expects it unless configured otherwise). Other than that, it's easier to support than a bunch of desktop Office installs. I think people will be pleasantly surprised when they see just how far open alternatives have advanced.
If readers are wondering, "Will this work for us?"
Yeah, it probably will.
Once you associatee a pictogram with a given function, finding the exact button you want with a pictogram is easier than finding a text only button.
The best interfaces were done in the late 90s, when we had combined text + icon buttons.
The text helped with discoverability when youi were not yet familiarized witht he program, while the icons helped to locate the buttons faster once you familiarized yourself with it
>” The best interfaces were done in the late 90s, when we had combined text + icon buttons.”
All based on the building blocks laid before anyone had heard of a Macintosh or Windows and vendors squabbled over trash cans and four colour flags..
UI/UX really started to go downhill when the design school idiots got control and decided they knew better…
>” you just say what it is. File. Copy. Paste. Select. Words.”
Also a very important consideration for over the telephone (or across the desk partition) support.
Whilst you can use, if you remember, the keyboard shortcuts, they are of zero help to the user as without the visual cues they will simply treat them as “magic incantations” and so fail to learn…
I gave OnlyOffice a try a few weeks back. Someone had recommended it to me, suggesting that it might solve a few compatibility issues I was having with xlsx files in Libreoffice.
I wasn't impressed, and its compatibility didn't seem any better.
Apart from the ribbon interface I found it disturbingly sluggish on what is otherwise an adequately performant Core i5 laptop - guess the fact it is pointlessly running through a browser engine answers that question.
I also very much hope that Libreoffice doesn't drop their traditional interface just because Microslop has insisted on fscking up the Office interface since 2007....
The extent to which the "whole team can collaborate" seems a bit limited by it being a single-user app confined to a single workstation.
Maybe I'm misunderstanding things about the existing Collabora offering, but I thought the main point was that with the documents being hosted on the server, it allowed for _simultaneous_ collaboration between multiple users. Maybe concurrent editing wasn't part of the spec (I'm not gonna go diving to find out) but just the fact of shared, mediated access to a central repository of documents would have made that a boon for collaborative work.
Does this new version actually support multiple clients, like over the LAN? If so, then I suppose as a version that's easy to install on the desktop without needing to figure out the rest of the server stack to host it, that could be handy. But if it's just the server in a box, for one local user, seems a bit daft when there's LO to provide the same thing.
Or is what they've done with the UI just so desirable that it still beats out OG LO even with this overwrought architecture?
@Nyle
The post above yours was on about OnlyOffice not OpenOffice (aka parrot, deceased*).
(*) I still use Apache OpenOffice 4.1.x under Linux on an ancient core duo Thinkpad with a nice keyboard and a relatively bright if low resolution screen. Needs less resources, especially graphics card related stuff, than LibreOffice local install. I also know where the bugs are.
I can only assume that by pushing the UI into a web/browser technology that the LO UI development can be offloaded on to external browser development projects. This might be driven by a graying and shrinking pool of LO developers. (I don''t know.)
Also might be driven by the declining specs in (real terms) of lower end PCs that are likely to dominate the market for some years. Running the fully local version of LO on the 2026·28 version of the "Netbook" or 4G/Celeron Chromebook might an unpleasant experience.
Or could be an outbreak of the brain liquifying daftness virus from which we appear to be suffering an epidemic.
As I have a local web server for an AL yum repo, I could test whether there is much difference between the local and web versions preferably using the same storage. I have an old, ca 2015, reflashed HP Chromebook (4G/Celeron) that now runs a lightweight Linux distro which might be a decent client side test machine. It ran Win10 like a decrepit tortoise.
Having a "web ui" is perfectly fine for a online version, where you have a back end and a front end. Having it on a local version is useless bloating and inefficiency. Like using containers or flatpacks for everything, for example. It's the endless bloating of software caused by the fact that hardware was cheap. It's no more. It's time to rethink software. But no one will do it.
So, why don't we just add AI? I'm sure it will sell better with AI. (sad sarcasm)
"I can only assume that by pushing the UI into a web/browser technology that the LO UI development can be offloaded on to external browser development projects."
The original Netscape Communicator and its current incarnation, SeaMonkey sort of used this approach for the UI of the mail/news client but that wasn't altogether surprising as the application was (and is) a combined browser and mail client.
What's slightly more surprising is that Thunderbird continues with this, using an internal browser. Perhaps Collabora could look at some sort of merge because there really is a need to combine the office suite functionality with mail/calendar/etc functionality. I believe that early versions of Open Office included Seamonkey code for a mail-merge function. It's a pity it was never surfaced as part of the overall application.
> Why? Just why? There is an existing local alternative that doesn't need this.
Multiple reasons.
1. Because they could.
2. Because it works for the competition: it works for OnlyOffice, it works for all those Electron apps from Slack to Signal.
3. Because this way they can distribute a local version that doesn't need a web server.
4. Because you can use it offline.
I am not saying that they are especially _good_ reasons, but they _are_ reasons.
But if you use COOL and want a local version, this way, with relatively minor code changes, Collabora could offer a version that makes it possible to work with no internet.
I suppose I'm looking at it from the user's point of view. There is a local version that doesn't need a web server - it's called LibreOffice, no Electron style bloat. If I need a local version that's what I use. If I needed online collaborative office collaboration I'd use the online version. But then I'm not looking to sell something for local use and trying to persuade customers to not look at things that way.
What would really be cool, if not COOL, would be to get together with Tbird and make a combined app. That might even be worth buying.
I think there are good reasons for making an online version of LO. If you *really* want to promote LO as an alternative to MS Office to businesses, you need it these days. It's disjointed to say use 'this' on the desktop, and 'that' online. I have no idea how many documents are created in Google Docs, but I bet it's billions.
Exactly this:
1. Our team needs to collaborate, but we won't waste time emailing various versions of files back and forth, then trying to merge multiple sets of edits. When one of us makes a change, everyone can see the edit immediately.
2. We have no interest allowing Microsoft or Google to train their AI on our work product.
Office365 could be completely free and we still wouldn't use it. No way in hell we'll let that exfiltrate private data.
It’s also disjointed if the desktop version can’t be used in place of the online version ie. The user should be able to use either clients and switch between online and offline working depending on the document being edited.
Basically, think of the offline client as download and install once online client, that doesn’t need the added memory bloat of the browser.
> But if you use COOL and want a local version, this way, with relatively minor code changes, Collabora could offer a version that makes it possible to work with no internet.
This should be on the list regardless. One big problem with it is the need for constant Internet access. Sometimes the Internet is out. Sometimes you're somewhere which doesn't have it.
Only network features need to be suspended when the Internet goes out. That should be fine, albeit with some minor versioning friction if multiple users were offline editing the same document at the same time.
I've used Collabera online on an iPad. If all you want to do is to edit a single cell in a spreadsheet you created using LibreOffice on a Mac and then saved in iCloud then Collabera is just about OK. It wlll still drive you mad and it might take a few goes to select and edit the cell but with perseverance you can make that edit, have a go at saving the file and breathe a huge sigh of relief that it only took ten minutes this time.
I run Collabora in a container on my Nextcloud instance. It works fine for my needs (spreadsheets mainly).
In a worst-case scenario, if the containers fall over I can rysnc the actual .ods over and edit it locally. It's haven't needed to, but it works OK.
There was some faff in its configuration initially, but if LibreOffice can simplify the installation and integration, they'll be on to a winner.
Done properly, this could work well. Desktop or cloud, who cares - get the basics for free, or pay for the proprietary bolt-ons. Hope we can get away from any "That platform is yours, this one is mine" mentality. Would require a browser-based local UI option for the free version, but not impossible once the smoke has cleared.
I used to use LibreOffice, but these days they feel more and more irrelevant. The desktop versions could all use a major UI overhaul, an online version is a decent idea, but not necessary, but they need a solid mobile version and a way to sync the documents between them. Needing the mobile versions, I'd basically left them behind and tried the other well known office suites. Ended up using Apple's products because they do what I need and aren't MS or Google. Lesser of the evils I guess? I would love to see LibreOffice become more relevant again.
Online is now the use case.
Remember how things were in the dark ages? Emailing a file to a teammate and waiting for them to finish their edits before you could start more? Or having to track changes and manually fix what's out of sync?
Office software is commodity at this point. The only thing which takes more market share from Microsoft is collaboration tools...for all those of us who don't want a Microsoft account to use our own f'ing computer.
I could have posted this as a reply to several other posts but I thought I'd branch out on my own.
Firstly I think there is room for all the options. I've never used COOL but I have a copy of CODE running in my HomeLab. The issue I have with both is that they run best in conjunction with a "cloud" stack. The recommended deployment of CODE is with Nextcloud. That's a pain because my preferred cloud stack is Owncloud so I end up running both wgich is wasteful.
I can see CODA being useful if you use COOL as a business but you need to have offline capability in some instances. With CODA you have essentially the same UI. Remember that the end users may be members of the general public not IT guys
I'll say right now that I've never liked the ribbon interface. All the office suites I've ever used right back to WordPerfect 5.1 for DOS and StarOffice on an under powered second hand laotop running Linux had proper menus, so I use LibreOffice fir offline work.
Now as for LOOL as far as I recall it didnt force a ribbon UI on you and most importantly it was agnostic about which web server you used (although the reference install used Apache). So there's a use case for on prem deployments that dont want to go to the trouble of running a cloud stack. So COOL when you need the collaborative stuff that comes with Nextcloud and LOOL when you just want to run as much as possible server side and simplify the desktop build. If they keep the ability to run in any vanilla web server then pick your favourite HTTP daemon.
I'd like to begin by saying that, much the author here, I don't really go into the "ribbon vs. non-ribbon" debate that starts anytime the issue of office suites comes up. Both in LO and in Collabora Office, I remove toolbars and use keyboard shortcuts (and, of course, the Side Bar/Navigator).
When Collabora Office came up with their new* offline version, what troubled me were the deeper implications. There has been a long-standing feud between The Document Foundation and its "ecosystem" companies: enough so to make for a rather funny office sitcom. The trouble is, as of late, this "ecosystem" has been reduced to a single company: Collabora Online. Collabora Online has also concentrated the lion's share of code contributions to LibreOffice (as well as most active developers from back in the times of OpenOffice.org: veterans such as Caolan McNamara, Lazslo Nemeth,Thorsten Behrens, and many others.)
Tensions are on the rise. Both the release of Collabora's offline Electron version of the suite, and now the de-atticization of the Libre Office Online project are broadsides in an escalating conflict. I'd like to be optimistic, but this could well lead to the Collabora Online "soft fork" from TDF becoming a "hard fork".
And I just don't think it's the right time for yet another office suite war, to be honest.
* Collabora Online have been offering an offline version of their suite for quite some time now: it was simply a rebadged version of LibreOffice. I think that has become "Collabora Office Classic" now, or something to that effect.
>” but this could well lead to the Collabora Online "soft fork" from TDF becoming a "hard fork".”
I suspect , from the little I’ve read about the tensions, it seems the new TDF wants to be able to walk away from Collabora with a source code base it has full control over.
TDF can certainly walk away with the codebase, but they'll lose the developer expertise. Also, it appears that TDF, as a non-profit, can't really hire developers directly. So they'll be left with community contributions, or they'll have to set up a company to hire developers, and then the whole cycle start again...
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2026
