LibreOffice: The Next Five Years | Hacker News Reader

LibreOffice: The Next Five Years

lwn.net

412 pointsingve425 comments

It's weird to me that people complain over "old and ugly" interfaces. Consistency is worth so much more than fad-chasing. Learning something is more worthwhile if you are able to trust that it won't change underneath you.

Take MS Office for example, I was made to learn how to use it in elementary school, we had an exam where we got a printed out page and had to reproduce it in an hour. This was supposed to be fundamental computer knowledge like learning to type.

I still have the same keyboard layout as back then but MS Office now has some janky "ribbon interface" which bears no resemblance to how it used to be. Although it should be criminal for public schools to teach proprietary software, we can forget about that for a minute and instead consider how futile it is to teach things that are not open standards or at least free software. You have absolutely no assurance that this knowledge will still be applicable even just a few years later.

My hope is that free software projects will attempt to preserve old interfaces (making them accessible via initial configuration) when they make updates. Besides, you shouldn't be replacing your GUI if your architecture hasn't got a clear separation between presentation and core logic.

Your school was wrong to teach any program. Ui follows fads and always will. Thus you need to learn how to use any ui that might come.

I agree with this, but it was far less clear in the 90s. I think a lot of people (myself included) were expecting/hoping that UIs in productivity software were converging on some sort of finished design or standard that there wouldn't be much reason to monkey around with.

MS Office programs - and many other proprietary software titles - were taught not only in many schools, but also in various vocational programs. You could get 'certified' in having 'mastered' a particular UI, and you expected that knowledge to continue to be useful for years. Seems quaint now, but there were some nice things about it. A carpenter doesn't expect their saws & hammers to radically morph every product cycle.

In the 90s word was still trying to take the crown from word perfect. Everyone knew guis were the future, but the F Keys in word perfect were important to know.

I don't think so, but I also don't think that it's going to be popular opinion around foss/computer people

MS Office software is used around shittton of workplaces and from time to time in day-to-day life, so making people somewhat proficent at it is good for everybody except MS competitors.

I'm not sure what do you mean by teaching "UI" - they teach how to get things done with given software.

It’s hard. I learned a lot of programs (some that don’t exist) autocad and lotus 123. Some of the principals of spreadsheets and vector programs I remember to this day. But having hands on with the software certainly helped learn it.

I used to feel that way about the ribbon, but I quickly got used to it and now consider it an improvement.

What I’ve never got used to is the flattening of the UI in versions 2013+.

I never loved the ribbon and I suspect that Microsoft underestimated how disruptive the change was for a lot of people. However, I saw a presentation by one of the lead designers around the time that it was introduced and he certainly made a good case for how the older interface had accumulated way too much cruft.

ADDED: Ah, this is probably it: https://www.youtube.com/watch?v=Tl9kD693ie4

(That said, MS Office in general had been loaded with so many functions that 1% of users used but that were absolutely essential to some 1% that it's something of a mess anyway. It's one reason I prefer the relative simplicity of GSuite these days.)

> I saw a presentation by one of the lead designers around the time that it was introduced and he certainly made a good case for how the older interface had accumulated way too much cruft.

I have also seen well-intentioned arguments for why Clippy was a good idea … but he wasn't.

To this day I miss the Word 5.1 on Macintosh System 7. That was more-or-less the Platonic ideal of a word processor; it's been all downhill since.

Clippy was literally the precursor of things like Alexa and Siri today. The technology wasn't there to back it, but the idea of an anthropomorphic assistant has literally come around to the mainstream today.

Recently, I installed Office 97 onto Wine just for curiosity.

I could still do much with it. The only feature I really missed from Word was academic references. I could create documents indistinguishable from modern Word using the old drawing features combined with modern fonts.

The drop down menus, combined with the floating toolbars were much more intuitive than clicking up and down some "ribbon". The drop down menus map directly to the logical path of a task in my mind, eg "Insert -> Image -> From File". Having a small toolbar at the bottom with basic graphical tools is much more convenient than clicking back and forth from tab to tab.

It's a shame MS couldn't accept that what they had until Office 07 was a workflow polished near to perfection, perfected after years of refinement and familiarity.

>perfected after years of refinement and familiarity

That can be a problem though. One person's familiar tool with its well-known quirks and peculiarities is a newcomer's inscrutable and illogical interface.

Generally, though, you shouldn't optimize GUI:s for new users but for intermediate users.

[0] Cooper, "About Face: The Essentials of User Interface Design"

That should be a tenant of UI design, especially for any kind of productivity software.

Think of the Blender, GIMP or Sony Vegas GUIs. Daunting the first time you open them, but after you've successfully finished a few creations you'll begin to appreciate their power.

For any serious work, overly-simplified GUIs are quickly exhausted. They may get a pass for IM or social media apps, but I can't think of many other useful applications.

You're thinking like an SV webdev who's largest concern is to onboard as many new users as possible so you can flip your startup.

Office is a professional productivity tool that's been around for decades and is used by millions of people for their entire workday, every day.

Learning to use a new tool takes a tiny fraction of the time you're going to be using that tool. One should make the tool best at actually doing what it's going to be used to do.

I'm neither in SV nor a webdev. I just remember how many ordinary users hated the ribbon when it first came out. Presumably they adapted over time; I can't really speak to whether Microsoft would have been better evolving the existing menu system. Personally, I could take it or leave it--and at least intellectually the arguments for the ribbon made sense to me.

Ordinary users hate change. They always do, regardless of how successful (or not) the changes are later judged to be.

This doesn't argue against change so much as argue that the costs need to be worth the disruption - you don't move houses over a carpet stain, and you don't rebuild your UI lightly, either.

> eg "Insert -> Image -> From File".

In the latest version of word the progression is:

"Insert > Pictures > From This device"

So, it's catering to tablet and smartphone users rather than desktop users, who have many "devices" on their system. :-(

I went through a recent experiment like this as well. Instead I installed Office 2003(The last of the old toolbar offices) on a clean Windows 10 system. I was looking to validate my theory about the speed and bloat of the new versions of Office. I just had this nagging feeling that the software wasn't as fast as old versions. It was confirmed. Using the old Office just felt so pleasant with how fast it loaded and simple it is in terms of cognitive load.

The ribbon interface was adopted in 2007. That's 13 years ago.

What is the jank? It's a tabbed interface of buttons instead of menus and sub menus.

The most common operations are descriptive buttons.

Give someone word 2003 or word 2020 who has never used a word processor before. Which one would be more intuitive?

It's bad UI if people just learn where to click. They should be able to think in terms of "I want to do X" and be able to do that as quick as possible.

Of course with specialised tools that's a different story. AutoCAD hasn't really changed its interface but it got superceded in areas by new applications that break people's workflows in a lot of cases.

It's just a word processor. The knowledge you should have gained when being taught was the concepts of what you can do with one.

That should allow you to adapt to different software. Once you've used a word processor you don't need to read the help page for Google Docs.

It's not like switching from Maya to Blender.

I don't agree on the free software projects point. Often they're made by volunteers. And anything too "daring" would be shouted down by the community.

I'm going to disagree here.

Alice is a new user who has never used Word before. Between word 2003 and 2020, I grant she'll pick up 2020 much more quickly - point in favour of the new one.

Bob is an employee who uses Word just short of 4 hours a day, 5 days a week writing and editing reports (the rest of the time he spends in Outlook, excel etc.). If there's a task that Bob does 50 times in a typical week, then it doesn't matter so much to him whether it takes a bit longer to learn the command (after about the 150th time he'll have got used to it), but it matters a lot how fast he can do it once he's learnt it. Bob literally becomes more productive by having an interface which he can operate by muscle memory, in a way that more than pays off the initial training costs.

Think of keyboard shortcuts for example: they're completely unintuitive to a newcomer, but with experience, Control+Z and Control+C, Control+V and the like save time, and time is money. My favourite word trick in this category, incidentally, is Control+Space "remove formatting" for text you've pasted in from elsewhere; it doesn't work all of the time though. You used to be able to do Control+Shift+V for "paste as plain text", I don't know which version removed that again but I consider it a great loss. Paste -> Keep Text Only takes just longer enough to be annoying.

So what's kept Bob from committing the ribbon interface to memory in the past 17 years?

The ribbon interface removed customizability, and inflated the size of all the common commands (and also made it impossible to get them back without 3rd party extensions).

Pre-Ribbon I could have the menus, file operations, font and paragraph settings, and the reviewing tool bar at the top of my screen. I could have the object and image editing at the bottom.

Post-ribbon the exact same amount of space is does a third as much. Reviewing for some reason is now on a separate tab. The menus are less commonly used functions are non-existent, stuck somewhere under "File" which now takes over the whole screen when I open it.

It is not just "different" it's worse. It removes the basic ability to prioritize my interface to the types of tasks I'm doing, in favor of some vaguely defined every-user who is not actually a real user.

Just to be clear, the ability to customize existing tabs, and make new ones with whatever commands you want on them, is insufficient? Or is this only a complaint about the defaults?

For those who don’t know: https://support.microsoft.com/en-us/office/customize-the-rib...:

“What you can customize: You can personalize your ribbon to arrange tabs and commands in the order you want them, hide or unhide your ribbon, and hide those commands you use less often. Also, you can export or import a customized ribbon.”

Thank you! I didn’t even know this was possible.

They've optimized it for the least common denominator, not for specialized use or for power users. Short term this is a positive approach as it's more approachable for people who are just learning or are very lightweight users. But for people who have a lot of experience or use it often, it's a serious regression.

I'm a "Power User" and I've memorized every single hotkey on the Excel ribbon interface to the point that I don't care where they are.

Wrap text? Alt H A W

Open settings? Alt T F T

Change to page break view? Alt W I. Go back to normal view with Alt W L

Take the border I currently have on the top edge of a cell and swap it to the bottom? Alt H B M Alt+T Alt+B

Paste values transposing? Alt H V S V E. Want to follow the old accelerator? Alt E S V E also works

Power users don't need to care what the ribbon / menu looks like. They learn hotkeys.

I'm an emacs user. You have made me feel warm inside.

Emacs is a lot like Excel in many ways: they both make you feel like wizards. I suck at excel, but I've seen some analysts who are crazy good at it. You watch them use it and numbers and shit just start appearing out of nowhere. It's pure magic!

On top of that, old shortcuts still work. For menus that disappeared more than a decade ago!!

power users have macros and hotkeys. I've met plenty of excel power users, and I haven't met many excel power users that click on anything in the ribbon all that often. Maybe it's different since they're bankers and the amount of time they're spending on spreadsheets is an order of magnitude more than other people that could still be called power users but maybe don't have their lives revolve around excel and powerpoint, but hotkeys and macros are where its at

Except this is where the intermediate user point becomes valuable: if you spend close to all your time in spreadsheets, then yeah, sure, you're probably in hot key land.

Of course that's discounting the idea that a power user would still want to customize their UI to get less frequently used but still common functions into view and keep them there.

Which is what I, an intermediate user, definitely want: at the time the ribbon happened I was in chemistry research. Excel and Word are important parts of that process! And having them work the way I want was valuable to me. Just not so valuable that I needed to learn a huge number of hotkeys. And yet somehow, the ribbon manages to still be in the way rather then letting me keep a productive, dense UI in view for when I am in those tools - which I'm definitely not in all the time since most of the actual job is being in the lab.

The Formatting/Reviewing split was one of the most irritating possible things as a result of the ribbon: I can't keep the formatting tools in view alongside the Reviewing tools? Like, what?

I believe it's because the ribbon is context-sensitive. What shows up depends on what has focus, and that makes it harder for retention, similar to how Office 2000 has automatically hidden menu items that ended up being confusing.

At least for this Bob here:

1. The ribbon interface has more levels of depth, which equals more clicks. If most of your commands are in the same toolbar, you can have a lot of commands that are only one click away, even if you don't customise. For example, to "insert symbol" in word to get an em dash or non-breaking space or similar, you have to select the insert tab, click the symbol dropdown and then "more symbols ...". Used to be one click on the "omega" button, or two via the menu (insert / symbol). The fact that I have to move the mouse from one end of the screen to the other and then half the way back again also doesn't help.

2. Maybe this is just me but Alt+N, U, M for "insert symbol" is a lot harder for me to memorise than "the button with an omega up there to the right". It feels less intuitive somehow - I really like working with visual muscle memory (if that's a thing) where I remember the place and shape of a button.

3. Customisability. Yes you can do this to some extent with the ribbon, but in Excel you can't just swap out "merge and center" with "merge across (but do not center)", you'd have to replace the whole "section" - I've tried.

4. Speaking of visual muscle memory - the number of times I've come in to work and found that my ribbon has changed in an update overnight. Some of these are work installing new plugins, but most of it is microsoft adding new "features". No thanks, I don't want my documents "integrated with LinkedIn" when I'm working to a deadline.

5. There have been several major changes to how the ribbon looks and works as Word has gone through new versions since it was introduced - in a lot less than 17 years. Of course that happened with toolbars too (I think around 2000 everything went "flat style", but you could at least turn it off), but I found it easier to adapt back then. Maybe I'm just getting old but each new version that comes out feels like learning half the things I need for work all over again.

I have better luck pressing shift as the first button in the sequence for the "paste as plain text" but I agree it does not always work.

Shift+Control+V

indeed. Focusing only on the newcomer and more precisely the person using it for the first time is just a fraction of the "problem".

Your argument supports his - Ribbon is better for discovery, keyboard shortcuts are there for routine performance.

It's not just begginers who value feature discovery.

> used to be able to do Control+Shift+V for "paste as plain text", I don't know which version removed that again

Try Control+Alt+V. I have Office professional plus 2016, works here, however it opens a popup. Plain text is at the end of the list, making the complete sequence Control+Alt+V, End, Enter.

That is no shortcut, that is an abomination. Basic Ctrl-V should be simple plain text paste. honestly, how often do you want the formatting from the source, vs. seamlessly matching the target?

> That is no shortcut, that is an abomination

I can see how it's unusable on a laptop, on a full-size PC keyboard that's OK.

> how often do you want the formatting from the source, vs. seamlessly matching the target?

I don't write word documents unless I want rich text features. Just for text there's nothing wrong with plain text or markdown files.

Copy-pasting from visual studio is the easiest way to apply good quality syntax highlighting to a code snippet. The IDE is even smart enough to copy them with dark-on-white colors despite I use dark theme.

Also, I often cut & paste within the same document.

>> I don't write word documents unless I want rich text features.

Just for clarification, Paste as Plain Text doesn't paste an unformatted string, it blends with the target formatting. So for example if you are copying a phrase you copied off a web page into a Word doc where the current paragraph is 12pt Georgia, you get 12pt Georgia no matter what the formatting was on the original document. The default Paste will attempt to replicate the color, weight, font size and family of the original web page into your Word doc, which is highly unlikely to fit in with the rest of your document.

Thanks!

I remember in years gone by, there was a (Sun?) keyboard with extra keys down the left with names like cut, copy, paste, insert, close and so on. I'm half tempted to buy one of those gaming keyboards with programmable macro keys and map one of them to this "paste as plain text" sequence.

Speaking only for myself, but the ribbon interface helped me learn the majority of keyboard shortcuts I use today.

Once I leveled up past basic Ctrl+C/V/X, hitting Alt and having the Ribbon UI guide me through the shortcut combinations was helpful. I concede Control+Shift+V is easier, but having a visual guide until I finally memorize all the different paste options with ALT H V has it's merits as well.

Paste as Plain Text should be the default. The inability to do so via the right-click context menu in Office (like Chrome allows) drives me nuts. Perhaps an exception if you are copying/pasting from the same document, although personally I'd still prefer plain text.

You can make it a default. It's under 'Advanced' in options.

Also, when you paste, Word shows a small button labeled 'Ctrl' — you can click it or press Ctrl, and adjust your paste.

As an ex trader, I can confirm that the new ribbon interface, when it first came out, caused me and my colleagues a world of pain.

We knew all the shortcuts keys in the old version. The ribbon UI meant that a big chunk of the previous shortcuts didn't work anymore.

Killed our productivity for a good few months.

Had the ribbon interface been the original UI and the traditional menu-based interface the new UI, it sounds like you and your colleagues would have still faced the same frustrations. To me, it sounds like your issue is with change rather than the effectiveness of either type of UI.

No, not really. Some 1 or 2 key shortcuts became 5 key shortcuts with the new ribbon UI.

It was a real nightmare! Especially when the old shortcuts were also remapped to a completely different function, and your moving s so fast you don't realise where your data has gone!

The ribbon allows you to add your most common used commands to a quick shortcut list that are only 2 keys.

For me, that's the point.

For a program as ubiquitous as Word (and as complicated as Word), unless there were clear benefits from the new interface, it shouldn't have been changed. The change itself was the problem.

Your comment is getting at an important, fundamental UX principle: all conceptual change is inherently painful, and every time you ask a user to re-learn something you are always wasting their time, even if the new design is better.

Sometimes a UX designer looks at two interfaces as completely separate, and thinks their job is just to pick the best one. The reality is that usually UX decisions can't be separated from the current state of the app. UX design is more like surgery, and the invasiveness of the operation has to be a consideration.

Having said that, I'm mildly skeptical that the interface didn't need to be changed. The old interface for Word was extremely user-unfriendly. If you go through a few of the interviews about Ribbon[0], there was some really interesting thought put into the new approach around discoverability, and while I'm not sure I agree with all of the theory they used to justify it[1], the changes did seem to be addressing a real problem.

But I think it's debatable whether or not the discoverability benefits were worth the pain, and I think there may have been better ways to roll out the changes.

[0]: sorry, too lazy to hunt them down and link them

[1]: I am mildly skeptical of contextual menus/toolbars. They're not wrong, they just have drawbacks that people don't always consider. Sometimes it's useful to tell people what is disabled, and to give them grounded, unchanging "landmarks" in the UX that they can use orient themselves while using a program.

I think that choices like this come from a weird symbiotic relationship between UI designers and product managers. The UI people need to justify their jobs; if they sit on their hands all day because the UI works well, they can't make that justification. Managers don't want to "see a product stagnate" and to fix that they want a "fresh new look". The users don't enter into this relationship.

Of course there's an issue with change. That's the whole point. If you were a pro Emacs/Vi user that used shortcuts to do pretty much everything and it made you fast and productive, what do you think would happen if suddenly you had to use new shortcuts for all the ones you had already commited to memory?

Your experience reminds me of https://xkcd.com/1172/

But, the new ribbon interface was much more inconvenient.

If I recall correctly, pasting data while removing duplicates in the ribbon interface was Alt+AQORT, but only Alt+XY in the old Excel (I don't remember what X and Y were, but I know it was a 2 key combo!)

Excel kept the old Alt based keyboard shortcuts, e.g. Alt D F F for Data > Filter > AutoFilter still works.

If there was an alt-keyboard shortcut for paste removing duplicates, that should still work.

That's the point; the old shortcuts did not always work. Worse still, they did something totally different. For "power users", it was a horrible experience.

Good old Microsoft changing the shortscuts every major release, because. Because? Because fucking why? Why do they do this.

> The ribbon interface was adopted in 2007. That's 13 years ago.

Came here for this. The first version of Office for windows (which is what most people are familiar with in terms of UI) is from 1998, so the non-ribbon interface has been around for 9 years, vs 13 for the ribbon interface.

Office 95 doesn't look that functionally different from Office 2003:

Office 95: https://i0.wp.com/isoriver.com/wp-content/uploads/2020/01/06...

Office 2003: https://imag.malavida.com/mvimgbig/download-fs/office-2003-s...

Honestly even Excel 3.0 (1990) doesn't look that disimilar (https://goughlui.com/wp-content/uploads/2014/05/about-excel....)

So that's 17 years for toolbars vs 13 for ribbon

Still, I guess more MS Office users will have used the new interface for longer than the old one than vice versa (I googled for data, but couldn’t find numbers)

It even wouldn’t fully surprise me if, a few years from now, there were more MS Office users who never used the pre-ribbon interface than who had used it.

> The first version of Office for windows [...] is from 1998. [...] 9 years, vs 13 for the ribbon interface.

You are not even close to right. Excel for Windows is from 1987 (20 years with non-ribbon menus), Word for Windows is from 1989 (18 years with non-ribbon menus). And of course the menu bar paradigm itself is older still.

you came here to be wrong?

> What is the jank? It's a tabbed interface of buttons instead of menus and sub menus.

That keep changing and moving about.

I use Word on Mac. It has the ribbon. 90% of the time I use the old style menu bar at the top of the screen.

Word for Mac is almost the perfect evolution of the UI.

It has the menu bar, which every Mac app has since it's a fixed and global UI element. It has the ribbon which is good for many users and makes many tasks easily accessible.

I'm not sure how you'd make this compromise on Windows without making a mess of things, but on the Mac it works very well.

I've used office daily for 20 years but rarely have to use the ribbon menu at all and, if I do, it's rarely off whatever is up there by default. But every time it is, it's way more painful to explore compared to a plain text menu and after I find what I'm looking for, it's hard to get back to the default.

The ribbon interface was adopted in 2007. That's 13 years ago.

What is the jank? It's a tabbed interface of buttons instead of menus and sub menus

I've been using Office since the early 90's and the ribbon interface for the last 13 years. I still can't find stuff easily in the ribbon interface. It's like there's some cognitive barrier to me retaining the layout of anything beyond the Home ribbon (or tab?). Any time I switch to the other ribbons (tabs?) I find myself hunting, wondering if I've got the right tab.

There's something about the spatial layout that seems to impose an extra cognitive load. I know menus are spatial too, but it seemed easier to remember them. Perhaps, because each menu occupied its own area of the window when expanded. With the ribbons, the same area is reused when you switch tabs.

With Word and Outlook, I've got slightly better retention. However, with Excel, where I seem to have a wide distribution of seldom-used functionality, it's an exercise in frustration. I know something exists, but it takes ages to find it because the last time I used it was probably a few weeks ago.

The key question is: would somebody who learned to use office after 2007 with the ribbon interface appreciate moving back to the pre-2007 interface?

My guess is that more people learned to use office after 2007 than before, so the majority of people have no problems using the ribbon.

I still yearn for ms-word on dos, and oracle forms on terminal interfaces, once you learned the keyboard shortcuts, nothing was faster. Kind of like knowing vi(m).

My guess is that more people learned to use office after 2007 than before, so the majority of people have no problems using the ribbon.

Or, perhaps they are just used to a lower level of productivity? I've had 13 years to adapt, and I still find it frustrating.

On the other hand I found it instantly more productive, even right after it was introduced. No longer did I have to remember which menu an option was in. It was all right there.

Not sure why this got downvoted — a personal anecdote as a reply to personal anecdote.

On the subject — I’ve learned the “classic” office but really like the ribbon. I’m not a power user though — just a doc or a spreadsheet every now and then.

It's not all right there - some things are under another ribbon tab, which doesn't seem all that different from putting things in a collection of menus. (Except that visually it's now a jumbled mess instead of a clean and easy-to-scan linear list.)

> Except that visually it's now a jumbled mess instead of a clean and easy-to-scan linear list

I think I find the "jumbled mess" easier to scan because my memory is visual, and the items are visually distinct. On the other hand, with a list I actually have to read every item.

> My guess is that more people learned to use office after 2007 than before, so the majority of people have no problems using the ribbon.

Or maybe they didn't have other options...

MS Word 4.x on the Mac was pretty sweet.

I learned on the pre-2007 interface, and (at least for Word) I would go back to the Word for DOS keyboard-based interface including the little template you put on your keyboard.

I agree with this completely. It's a lot easier for me to read through a list of things in some category than it is for me to scan through a bunch of irregularly sized tiles for the icon and description of the configuration I need. For me, all of the pictures just lower the density of the information and cause it to take longer to parse every single time.

I also question the basic premise that the tabbed interface is somehow more intuitive for new users. Like many UI related assertions, absolutely no concrete evidence is being provided to support this point. I would think that it would be easiest for new users (as it is easiest for myself) to have options sorted into neat menus with uniformly sized elements where information is laid out in a vertical list (which is how lists are written in basically all left-to-right and right-to-left languages) so that it is easier to parse quickly.

> “I also question the basic premise that the tabbed interface is somehow more intuitive for new users. Like many UI related assertions, absolutely no concrete evidence is being provided to support this point.”

Uh, yeah there was? Go dig through the wayback machine for Jensen Harris’s blogpost series about the design of the ribbon, or watch his MIX08 talk on it; he was the design team lead for the change, and the reason they did it was because the old UI was failing their users. There were usability studies done during the ribbon design which fedback into, and changed, how it worked to adjust to ordinary people’s expectations and needs.

> “I would think that it would be easiest for new users to have options sorted into neat menus”

Like many UI related assertions, absolutely no concrete evidence is being provided to support this point.

"There's absolutely no evidence for this" "yes there was, here's what to look for" "downvote". HN.txt.

Top level menus full by Word 95. Office 97 added command bars. Nested menus and toolbars full by Office 2000. Office XP added Task panes, full of features by Office 2003. "the Task Pane was the last attempt to find a way to scale old-style UI to programs as full-featured as Office. Although it was a successful stop-gap measure, it ran its course in only two versions." - https://docs.microsoft.com/en-us/archive/blogs/jensenh/new-r...

> "The downside [of nested dropdown menus], however, was clear and eventually terminal: increased complexity. It's much more difficult for people to form a scanning strategy with hierarchical menus: you have to keep track at each moment which levels you've visited and which you've haven't. What was once a simple structure to visualize was now a more complicated, branching structure."

- https://docs.microsoft.com/en-us/archive/blogs/jensenh/ye-ol...

> "As we watched more people use the prototypes, we started to understand more the scanning process that was taking place. Later on, we did eye-tracking studies to watch how people scanned the Ribbon"

> "I was reading a blog entry of someone who was kind of critical and dismissive about what we're doing and our designs. One of his criticisms was "how bad the usability of the Ribbon would be because it's got icons scattered all over of various sizes." What we've learned is actually the opposite. People can scan disparate patterns more easily than homogenous patterns. When we use more toolbar-like layouts--a bunch of equally-spaced, equally-sized buttons, people scan them less quickly than when each chunk has a memorable layout. So we actually try explicitly to vary the layouts between chunks--it helps people find the thing they're looking for more quickly. That's something we wouldn't have known if we didn't have a commitment to watch people work."

- https://docs.microsoft.com/en-us/archive/blogs/jensenh/be-wi...

> "One of the concepts behind the Ribbon is that it's the one and only place to look for functionality in the product. If you want to look through Word 2003 to find an unfamiliar command, you need to look through 3 levels of hierarchical menus, open up 31 toolbars and peruse about 20 Task Panes. It's hard to formulate a "hunting" strategy to find the thing you're looking for because there's no logical path through all of the UI.

> Office "12" consolidates all of the entry points into one place: the Ribbon. So if you're trying to find a feature and don't know where it is, the scope of your search is drastically reduced. Click on the leftmost tab, and click across the tabs until you reach the end. That it. It's either there or it's not--there are no other "rocks" to look under, no other places we've hidden functionality. We've found in early tests that people find it easier to discover how to do new things in the Ribbon, and they're more apt to explore the UI looking for better ways to get things done."

Testing showed that people found the Ribbon easier, found more things, and were more willing to explore it. - https://docs.microsoft.com/en-us/archive/blogs/jensenh/enter...

> "* Back in the olden days of designing software at Microsoft (say, pre-2003), design decisions were mostly supported by guesswork. [...] How much data have we collected? [in the Customer Experience opt-in program] About 1.3 billion sessions since we shipped Office 2003 (each session contains all the data points over a certain fixed time period.) [...] one of the biggest reasons that we decided to do the new user interface for Office 12 is simply that, for the first time, we have the data we need to make intelligent decisions.*"

- https://docs.microsoft.com/en-us/archive/blogs/jensenh/insid...

And that's not even from the design blog, which appears to be offline and not in the wayback machine.

Totally wrong about people finding menus easier, totally wrong about there being no evidence based design behind the ribbon.

> It's either there or it's not--there are no other "rocks" to look under

Nit pick on this quote: every time I've used the ribbon, I need features that have been collapsed into a "more" list (hamburger menu?) for that section of the ribbon.

Also:

> you need to look through 3 levels of hierarchical menus, open up 31 toolbars and peruse about 20 Task Panes.

I'm really dubious that the ribbon is better in this regard. Those menus still exist, and I end up falling back to them with some regularity after searching the ribbons and not finding what I want (perhaps it is hidden in a hamburger menu). Furthermore, each option in the ribbon takes up far more screen real estate than an item in the drop down menu, and there are sub-dialogs still abound.

I really love good UI design and I'm convinced I don't have what it takes but I do keep seeing arguments and interfaces that seem quite weird to me. I would do it like this:

The desire to keep the document (or whatever it is) worked at on the screen during the "hunt" I consider a mistake. When mixing paint on a pallet the eyes are focused on what you are doing. You might want to look up to the canvas and back down again several times but there is never a need to do both.

I also feel drop down menus are a mistake. There (imho) should be a key on the keyboard to bring up the "hunting" screen (perhaps one that can be panned to the left right top and bottom with the arrow keys) and every "button" on that screen should be visually mapped to a key combination. F keys are great for this. First F key for the "button" group, second F key to pick one or a 3rd for 12 x 12 x 12 options.

When folder trees stop working because you have to much in them you should do tags (in addition) that can hold duplicates of the functionality. Then when tags also fail to scale you need a search feature. Each interface "button" or group should have a lot of hidden text by which it can be found. Typing cursive should highlight the italics "button".

It would definitely become a big mess but my gut says it can be sorted out and be equally accessible to people using it every day and first time or rarely users.

I’m an artist who works mostly digitally. When I am mixing colors I have the color dialog right there on the screen next to the art, with my changes being reflected live. How does this color work with the rest of the work? That’s what’s important.

The only reason traditional artists use a palette far away from their art for their paint is because that works in the physical world. Mixing my colors right next to my virtual canvas is a huge speed-up compared to painting a ton of thumbnails with different colors.

So much bull crap stuff. The fastest way to find something is to search for it.

And...search for a menu item is a 2nd class feature on the OS. I wish I could click a text field on the menu bar and find stuff. I wish the actions were indexed with the documentation. Oh—and I wish that the keyboard shortcut for that was standard. Not on my Mac, but I’m pretty sure that I had to type it in from a Mac productivity magazine.

> So much bull crap stuff. The fastest way to find something is to search for it.

Faster to search for "save" or press Ctrl+S? Faster to search for "save" or click the save icon at the top of the screen? Searching is slower, clearly. Even if searching was always faster, it doesn't refute "bull crap stuff" because there was no claim that the Ribbon is the fastest full stop, it's a balance of discoverable and usable; search requires you to know the right search terms to use, and it suffers the problem of the old Office self-customizing menus - the results aren't always in the same predictable order - and if what you want is not in the top few results, you don't know if the feature doesn't exist, does exist by another name, or is in the hundreds of results you aren't going to read. And it has no way to surface features and make them discoverable.

You search the first couple times you need something, then you memorize where it is in the menu. If you keep using it repeatedly, you'll learn the shortcut, since it's displayed prominently every time the menu drops down.

On the Mac search for a menu item is almost always available as the very first item in the "Help" menu, and the keyboard shortcut for it is both standard and fairly easily guessable: Command + ? (e.g., Command + Shift + / on a keyboard where "?" is the shifted version of "/").

Not every program implements a Help menu, and not every program that implements a Help menu implements the standard shortcut. But in my experience, a large majority do. (Nearly all native apps do, but even some Electron apps get this right. Visual Code does both, for instance, and Slack at least has the Help menu with the search field in the first place.)

I see a problem - target user group.

I'm watching MIX08 The Story of the Ribbon [1] and trying to be objective. He is optimizing for 50% who paste with mouse. Essentially it is Microsoft Bob 2.0.

Office was done, "good enough" but some people complained, there was bad press. Microsoft added intellimenus, task pane - acknowledged mistake. Then ribbon. It could work for new users. Not for me. Now I better understand why.

The menu / toolbar / shortcut UI provides progression - from least often to muscle memory. Menu is hidden, fallback, toolbar provides customization, no need File > Save or Edit / Clipboard once I know shortcuts.

Ribbon is always expanded menu represented like task pane. They display text when icon is not enough. And group name ... because something. Information density is extremely low. Actual toolbar moved on the title bar.

Layout demo looks good. But as writer tool it is awful, clutter is still here, just behind a tab. Could be fixed with Firefox-like customization menu, not with their spaceship like-control.

I can see how he applied design tenets but in very specific meaning:

* a person's focus should be on their content - constantly changing UI, "content" is styling

* reduce the number of choices - no way to customize toolbar, select one of the tabs

* increase efficiency - of those who struggle with discoverability

* embrace consistency - among office, not OS

* give features a permanent home - even universally known

* straightforward is better than clever - no hidden controls, more visual clutter

* design tenets has to be religion - stick to decision, do not listen to users

Same tenets produce my layout, which is opposite.

[1] https://channel9.msdn.com/Events/MIX/MIX08/UX09

[2] http://sergeykish.com/side-by-side-no-decorations.png

There's never any shortage of cranky commenters on HN with eloquently written 4 paragraph posts about how some complicated interface from 1993 is the pinnacle of computing productivity, demanding peer reviewed studies to the contrary.

I mean, you could argue that some complicated interfaces like VIM's modal editing are the pinnacle of productivity for power users.

But power users and ordinary user's needs are world's apart, and the office suites have to cater to both of them. Thus the deep divide.

> It's a lot easier for me to read through a list of things in some category than it is for me to scan through a bunch of irregularly sized tiles for the icon and description of the configuration I need.

I think you hit on something here with the irregular size of the buttons, probably moreso than the icons. The arrangement makes it difficult to just read through all the options left to right, because they're all jumbled.

A common/recommended design pattern for older Mac apps is to have a top toolbar of large icons, which is actually quite similar to the Ribbon in some ways. But there, the options appear in a single horizontal row. https://i.ibb.co/wd6MR9c/Screen-Shot-2020-07-10-at-11-58-37-...

Not only are they jumbled, but they also change size/shape depending on the width of the window. A large button when full-screened becomes a small button when in a smaller window. A large button in a smaller window gets expanded to show all sub-options when maximized. There is no visual consistency for the same button.

That was a feature designed to fit more options on smaller screens. I personally like it and find it helpful.

It messes up any attempt to recollect of the locations of certain features in the bar. Some iterations of the ribbon even fold groups into menus under single buttons, so the exact path to feature then depends on the width of the window. This turns the ribbon into a pretty inconsistent user experience.

I don't recollect the location of features by their spacial location, I recollect them by their place in a hierarchical ordering. The exact path to a function is always preserved, as the keyboard shortcut for that item is defined by that path.

Ribbons (in Office, at least) change that very hierarchy depending on window width. For example, take the Home bar in Word: the "Editing" and "Style" groups collapse to buttons with submenus before the width of the window is reduced to half screen width on my computer. This affects keyboard navigation in the ribbon bar, too, of course.

This doesn’t actually change the hierarchy of commands, as evidenced by the fact the keyboard shortcuts stay constant. I appreciate hiding these first before Font, since I use Font more than Styles or Editing.

What hierarchy is supposedly staying constant? Keyboard shortcuts have no hierarchy.

Keyboard shortcuts for functions in the ribbon map directly to their placement in the ribbon hierarchy. Tab, section, subsection, etc. You press alt and then a series of one or more letters that define the path to that item.

> I think you hit on something here with the irregular size of the buttons, probably moreso than the icons.

This would also explain why other icon based UI's like the tool palette in gimp don't seem to suffer from the same problem. A grid of buttons is about as "programmer UI" as they come but once you learn where everything is it's quick and easy.

Vertical lists can also work better with the wider aspect ratios of modern monitors. There's tons of empty space on the left and right in a typical Word document, but MS decided to suck up limited vertical space instead

> It's a lot easier for me to read through a list of things in some category

this also creates a fixed 'path' cognitively since you are reading/naming the intermediate nodes on the way to any given task, and the intermediate nodes are consistently located in the same place spatially and in terms of the path. This is not the same as 'funky star button 2/3 of the way to the right and 1/3 down of the sub-scrolling subpanel of the customizable ribbon'

> Like many UI related assertions, absolutely no concrete evidence is being provided to support this point.

Microsoft did massive amounts of user testing, just because you don't know about it doesn't mean it's not there. This is a good starting point if you have about 20 spare hours: https://docs.microsoft.com/en-us/archive/blogs/jensenh/ They also backed their tests with a ton of data and metrics.

Testing and metrics can be wrong like in this case. A ton of data can be too much where we find patterns to reenforce what we want.

I'm not saying that you're wrong, but what Microsoft was doing is closer to science than what HN does every time an article somewhat related to the Ribbon is posted.

I'd bet on Microsoft's approach over these rants, every day of the week.

Each person who voices an opinion is a real value and a real data point. What Microsoft is doing may look like a science but they are not asking the right questions so they get misleading results.

Reminds me of when facebook pushed autoplaying ads. A/B metrics when through the roof. Sales never saw so much engagement. Every slapping each other on the back. Ask a user they didn't understand what was happening, there was a fear that was introduced around scrolling your own feed.

Science is great but only attempts to answers the questions you give it.

I recall the first time I used the new windows 8 UI. It was absolutely horrifying. There were 2 different control panels with different settings sometimes overlapping. The traditional UI in which you could start apps by clicking on desktop shortcuts and manage traditional windows but certain apps including the start menu put you seemingly randomly in the weirdo new app experience where the window took up the whole screen. This was also true of clicking a url link in a different app.

It was bar none the worst desktop interface that anyone has ever released for a computer.

You have asserted without evidence that MS's UI is driven by science and you are assuming that Microsoft's research is in fact any good. Based on consistently mediocre results I have reason to doubt that. To support your position I suggest you actually cite some research instead of assuming it has been done competently.

I read those blog articles as they came out, and bought into the same kool aid as you did, but when time came to use it, it just didn't work as well as menus.

The contextual toolbar is one of the better ideas though.

I remember back when Visual Studio 2012 shipped with black & white toolbar icons, and main menu in ALL CAPS. This was all backed by studies that supposedly showed how most users prefer it that way, according to the lengthy blog post about it.

Let's just say the community's reaction was not in line with those studies...

I am also the same, I have a complete block on how to find things, they just aren't where I remember even though as others have pointed out its 12 years since the ribbon appeared.

In 2020 there is a search feature (similar to VS Code) so there's really no need to have to find anything anyway.

Oh nice that will probably solve my dilemma on knowing what I want but not knowing where it is!

It sounds like the issue is that the layout isn't spatial. There isn't a 1:1 mapping between place and object. The same place can have multiple objects depending on what tab's selected.

Adding to this is the ribbon content is just too busy. It fused icons with labels and stuffed it in some uneven grid. The regular toolbar was more or less even grid, well, a row. The usual menu is also structured on quite uniform grid. It's mentally easier to follow straight lines, we read in straight lines, scroll in straight lines, well, typing too. Yet when it comes to ribbon, the adjustable grid layout makes one adopt an adjustable scan pattern.

I find this pattern quite distracting. Always feels like I'm staring at the old-time toolbar customization dialog, as if someone spilled buttons out of a toolbox onto the screen, and to make it more fun, also mixed them with labels...

Anyone remembers the initial requirement for the ribbon to use the specific font, as it was used for positioning and could not be customized? Not sure if it's still the case.

I wish MS promoted a style-based workflow/mindset more than the free-hand formatting. Perhaps, this would by itself eliminate the need for the busy toolbars, if only during style design.

Interesting, maybe old menus had some trivial adequate nature for most users and ribbon, even with all its worth doesn't tap into that. Personally I'm still incapable of deciding.. because I hated nested menus (too deep, too adhoc) but the ribbon is a bit less easy to remember than I'd hope. That said it's a good step toward keyboard shortcut memorizing.

One thing about the old menus is that they were extremely configurable. Like, you could not just hide or reorder items, but reorganize the hierarchy if you wanted. People mostly used that for toolbars, but Office menus are technically a kind of toolbar.

Some of that still lives on in Visual Studio, where it is even extended to context menus: https://docs.microsoft.com/en-us/visualstudio/ide/how-to-cus...

Were Office toolbars customizable to that extent ? Also I don't think it applies to most users .. they were rote learners (because never had the time or context to learn to use the computer as an aid but as a protocol to respect)

The ribbon is also extremely configurable. You can move or remove anything you want. There's usually no reason to, but it's an option.

There's also the Quick Access Toolbar, which is exactly like the toolbars of old if you pin it below the ribbon (not in the titlebar).

Yes, the ribbon is a futile exercise. I'm pretty sure 90% of the people don't know how to use it, and the ones who know don't remember where things are. In this sense, menus are much more successful interface, at least you know where things might be. And in fact, when I use MS Word I first try to find things in the menu, and only reach for the ribbon for items I already know from previous experience.

I don't understand this criticism. The layout seems logical to me. If you want to insert something, go to the "Insert" tab. If you want page design options, go to the "Design" tab. What's so inscrutable?

There are plenty of problems. In the Mac, for example, menus are discoverable because you can go to "help->seach" and type what you're looking for. This cannot be easily done with the ribbon. Moreover, once you have a clue that you're looking for an "insert" option, you still need to search around in the ribbon, since there is no logic for where the option might be hiding. A menu you can read from top to bottom, but a ribbon you need to scan from left to right and top to bottom, making it much harder to use.

"In the Mac, for example, menus are discoverable because you can go to "help->seach" and type what you're looking for. This cannot be easily done with the ribbon."

I can't speak to the Mac version, but on Windows there is a search box at the top that returns a result for whatever you want.

"A menu you can read from top to bottom, but a ribbon you need to scan from left to right and top to bottom, making it much harder to use."

No, menus are 2D as well. There's 1 vertical dimension, and then there's the hidden second dimension where you have to hover over each item to see what's inside. For example, in Google Docs under the Insert menu there is an item for "Headers and Footers", which opens a menu with two items: "headers" and "footers". The ribbon flattens this and just lays both out in the tab.

"making it much harder to use."

"Much" harder? Really? Honestly, I've used both interfaces extensively over the years and I've gotten plenty of work done in both. I just don't see what the big deal is about this little war that's been going on for 13 years, and claims of one being "much" harder than the other seem greatly exaggerated. Like... oh, you had to scan in two dimensions to find functions. Okay, how much does that really impact your productivity? How much of your time is really spent scanning a menu? Once you found it.... you can make a shortcut for it. I've got a shortcut for all the common tasks I use, and it's really no different than a custom toolbar from the pre-ribbon era.

No way, searching items in a ribbon is much harder than on a simple menu, even if you have subitems. It is the difference between reading text (fast) and trying to decipher pictures, specially when they are not well organized. Most people have trouble with the ribbon, I am not the only one who noticed this.

Again, “much” harder is quite the exaggeration. I mean, maybe for you, sure, but I’ve had plenty of time in the past 13 years to figure out where things are. If I’m ever genuinely searching for something I don’t scan the ribbon option by option, I use the search bar.

As far as “deciphering” pictures, there’s an association that builds over time. When I see the picture I make the association automatically. There’s research that shows people recognize pictures faster than they read words, since words all look visually similar and pictures have more cues to differentiate between them.

You are talking about the experience of someone who has learned the ribbon for a long time. Of course, if you always use the ribbon, it will be more familiar to you. Just like if you were taught Mandarin as a child it will be easier than speaking English. It has nothing to do with how easy it is to use these two different alternatives.

I didn’t have a hard time learning the ribbon. Going back to my original post, I just don’t see how it’s so inscrutable as others in these comments claim. Could it be that all these claims of the ribbon being a “much” harder are overblown? It honestly doesn’t seem so much different than a menu system to me, and again there are pros and cons for both. I think if you’re going to continue making this claim you should attempt to quantify it.

It may also be that claims that the ribbon is easy to use are overblown, and are result of familiarity instead of any true advantage. If you want to claim that the ribbon is easier to use you should also quantify it, otherwise your personal claim is just an anecdote.

I'm not saying it's easier to use. I've said multiple times in this thread they seem just about the same to me. They're just two ways of displaying a hierarchy of functions. One way hides a lot of it in menus and takes up less room, the other way lays it out on the screen and takes up more room. I get a lot of work done in either interface. There are pros and cons to both. In the end I can write documents in both. I personally find it easier to discover new functionality in the ribbon, but I feel at peak competency they difference is a wash.

I don't think anybody is bemoaning the loss of the old Office buttons. The buttons were always kinda bad.

The problem is that the ribbon also killed the menu bar. Menu bars are a standard, known quantity across every application, which clean easy semantics for keyboard access. On some OSes they even augment the menus with searchability, celebrating the power of plain text.

Alt-key shortcuts work on the ribbon and search is build in as alt+Q (at least in Excel and Word). I use both daily.

> It's bad UI if people just learn where to click.

That is literally the entire and sole purpose of a user interface.

Oh it’s bad to learn where just to click, I recently learned that’s how most people do it.

You might find this interesting - https://www.truity.com/blog/intuitives-guide-getting-along-s...

There was a discussion on HN about this a while back but most people learn computers by “if I want x I do y.” And guards just how they learn. Changing UI for them is catastrophic because they literally have to relearn -everything-. Op is probably a sensor and finds that stability tremendously useful.

This really helped me be less frustrated in seeing how people can’t just figure out technology on their own and how I can just mess with it until I figure out the ”language” the app is using to have users interact with it. Most people can’t do that.

> "There was a discussion on HN about this a while back but most people learn computers by “if I want x I do y.” And guards just how they learn. Changing UI for them is catastrophic because they literally have to relearn -everything-."

Yes, but it's surprising to see a large contingent of them on HN and other technical forums. The rapid change of software technologies all but mandates adaptability and willingness to learn.

There is a population of people who just hate the ribbon. In my experience, they are usually attorneys talking about word. It's the same argument that their attorney ancestors made about Wordperfect 5.1 keystroke commands.

Personally, I don't see the big ribbon issue. It's easy to search for arcane features in the help menu.

IMO, they really need a Mac where you get the top menu bar and a ribbon-like thing. Or some sort of gnarly keyboard mode that would become emacs for lawyers.

Attorney ancestors?

> Attorney ancestors?

GP probably meant something like "forerunners" or "predecessors"; some form of "those who came before" that doesn't necessarily imply a familial relationship.

It's just a cutesy way of saying a long time ago when all the lawyers used Wordperfect. (A word processing program that I hated almost as much as Wordstar.)

In some places, not so long ago :)

I imagine a lot of big law offices still have computers with WordPerfect on them so they can read old files if necessary.

WordPerfect is still maintained by Corel. The most recent release was in May.

They also maintain PaintShop Pro, WinDVD, and WinZip. I think they're running a retirement home for distinguished software products.

Corel is the Microfocus of client software!

The thing is I get why people hang on to COBOL or Attachmate. The company licensing WinZip is the puzzler to me! :)

They are still out there.

Last year we recruited an attorney from a firm that still uses WordPerfect for all their documents.

There are still law firms that require all their attorneys to use WordPerfect.

I suspect it might be dependent on whether someone is stronger visually or spatially vs textually.

I find ribbon interfaces highly irritating anywhere I find them, which is rarely - they haven't really caught on outside a couple of MS apps, AFAIK.

I haven't used Word much at all since the ribbon interface. Like others, for whatever reason the ribbon makes doesn't work for me, neither cognitively or via muscle-memory. The categorization doesn't help me predict where to find something (fails "new user" ease of use), nor does repeated use make it easier or faster. And of course I don't use it (or windows) much, so I don't think in Microsoft(tm) English, adding a minor layer of translation for some jargon.

I have to use Excel enough that I've finally rote-memorized where frequently used things are. But when I have to find some tool I don't know where is, it is faster to google for instructions than hunt around the application itself, so I do that first.

I'm sure it is fine if you live in MS-land. I don't, and it fails for this casual user both for discoverability and streamlined use.

The ribbon changed again in office 365.

In looks only. Functionally and layout-wise, it's pretty much identical.

AutoCAD went along a similar trajectory actually - from menus to toolbars to a ribbon. The difference is, you can still turn each of those modes on and off, including using two or all of them at once. Whereas Microsoft just kind of went "Okay y'all are gonna use a ribbon now."

Edit: Forgot the AutoCAD command line. You can still turn that on too, and type commands like it's the 80s. It's actually kind of a paragon of a mature product that maintains all its older interfaces.

Haven't used AutoCAD in anger for a while, but my default rest position on the keyboard is still fingers over Esc, thumb over space. I always preferred using the command line. Not out of some 'power user' power trip, but more for screen space and efficiency. Being able to use AutoLISP was pretty neat too.

I came in during the "toolbar" era, but ended up using the command line quite a bit. It was custom ACAD macros, and then AutoLISP, that got me back into programming after a long hiatus.

> It's bad UI if people just learn where to click.

A combat pilot would respectfully disagree. :)

Source: Diving pulling 5Gs canopy side down.

I think the old style menus were more functional for technical people and other experts, but the ribbon might be easier for new users. MSFT seems to have understood that trade off. It's interesting that the Metro design, which seems closer to the ribbon, didn't take off the same way.

But from functional perspective I do like the old interface better, and especially that I can pullright to get at things without having to make multiple clicks.

Yes, the ribbon interface was initially pushed on the world in 2007, and it's still as bad as ever.

AutoCAD went ribbon around 2010...

The ribbon was one of the worst things to ever hit Office. I remember it was met with universal horror when it first came out, but they were fully committed (apparently not having done ANY user testing) and insisted people would get used to it and prefer it because: we paid someone a ton of money who told us it's better.

Like many others, I STILL to this day have to google to figure out where things are with the ribbon interface, probably the least intuitive interface I've ever used.

> but they were fully committed (apparently having done ANY user testing)

It is a little hard to parse what you mean, but Microsoft did a quite substantial amount of research leading up to the ribbon. A quick search on the net will lead you to summaries of it.

Sorry, I left out a *not.

I vehemently disagree, in 2007 NOBODY liked the ribbon. My wife's company refused to upgrade until absolutely forced to because it was such an atrocity. The most glowing reviews I ever saw were "I guess I'll get used to it eventually" and/or someone who used office once a year and found navigation easier when the extent of their workflow was changing font sizes.

https://www.zdnet.com/article/oh-the-horror-why-is-microsoft...

https://answers.microsoft.com/en-us/msoffice/forum/all/turn-...

https://www.infoworld.com/article/2651076/microsoft-s-ribbon...

People hate change, that's not a reliable indicator. You could increase the quality immensely, most users would still feel attacked because they have to spend an extra effort to adapt.

At the risk of starting a fire here, I'd say a much better approach to that was unity's HUD. You could simply fuzzy-search through all the menu items. It made using GIMP delightful. :)

That is the one thing I miss most about Unity! It was such a sweet feature. (Also, Unity minimised on the vertical space it used, which was great for small laptop screens.) The new GNOME versions are much improved compared to a few years ago, but I do wish Canonical had stuck with Unity...

> The most common operations are descriptive buttons.

The most common operation is trying to decide which ribbon something is supposed to be on. And it's difficult to get straight even after more than a decade, if you're just an occasional MS-Office user.

I curse the ribbon. I still haven't found all of the 2003 features.

With teaching UI you breed the class of people who will shout the loudest when something changes. Teaching your parents that "the internet" is the blue icon in the taskbar will teach them to call you when microsoft decides to install chromium-edge on their machine because "the internet is gone". Don't teach anyone how to use a specific UI, teach the concepts that made the UI look the way it does. And with that anyone will be able to transfer that knowledge to a slightly different looking UI without much of a problem.

>Don't teach anyone how to use a specific UI, teach the concepts that made the UI look the way it does.

With all due respect, good luck with that. A lot of people aren't interested in learning about file systems or how DNS operates. They want to know what button to order from Amazon.

> A lot of people aren't interested in learning about file systems or how DNS operates. They want to know what button to order from Amazon.

That's like saying people aren't interested in learning how to use a phonebook or how to call directory assistance. They just want to know who to talk to so they can order what they want from Sears.

How so? I don't understand this analogy at all. UI/UX design changes frequently involves changing the appearance and possibly location and even functionality of familiar things. If we changed the lettering in a phonebook, or even merely on the cover, to be, say, klingon font, we shouldn't expect the typical user to reach for that phonebook, yet essentially what the GP is suggesting is that it's sufficient for a user to understand the "phonebook concept" and users can learn the implementation details trivially based on that.

> I don't understand this analogy at all. UI/UX design changes frequently involves changing the appearance and possibly location and even functionality of familiar things.

And that's the fundamental problem because that doesn't take into account people who are familiar with how a particular application works. For example, if you compare a tape recorder, VCR, a DVD player, and a streaming service where you can play, pause, forward or rewind, it's essentially the same interface and that has been the case since the '70s.

It's similar to dialing a phone with a touch-tone system versus dialing a number on a smartphone (other than having to press a call button). The only major change in the UI was when the transition between rotary dial to touch-tone took place. Automobiles are another example (placement of the brake, accelerator, shifter, turn signal stalk, headlamp controls, etc (though things do differ from model to model to some extent).

So why do we keep changing the interface of computer applications every so often such that proficient users have to relearn how to do things? The reason appears to be that we're chasing a goal of making the UI more intuitive so that someone who hasn't used it before can figure it out, but that never seems to happen.

But, if people just learn how to use the existing UI, then they can use the application and other applications like it because of a standard interface.

> If we changed the lettering in a phonebook, or even merely on the cover, to be, say, klingon font, we shouldn't expect the typical user to reach for that phonebook, yet essentially what the GP is suggesting is that it's sufficient for a user to understand the "phonebook concept" and users can learn the implementation details trivially based on that.

A more accurate analogy would be to change the order of the listings in the phonebook to start from most common names and end in least common ones instead of being in alphabetical order because of the belief it would help new users find the information they're looking for faster compared to the traditional interface.

And by and large people learn the bare minimum needed in order to do those things. Certainly most of us have a pretty vague notion of what all exactly happens behind the scenes when we dial a phone. If the call doesn't go through as planned we might have some idea that the cell reception is bad or something like that. But we're probably not in a position to debug what's wrong in any significant way.

That's exactly true. People don't care about the means, only the ends.

The older UIs had those concepts, but it wasn't stuff like "how DNS operates". It was stuff like, "every app that works with documents has a File menu with New, Open, and Save in it". Or, say, "if you want to see all available actions for something that's represented by an icon, right-click for context menu". I taught my mom like that, and she was amazingly productive at learning new apps. She didn't always find the most efficient way to do something, but she always found a way to do whatever she needed to do.

Modern UI ditched all that. Just about the only consistent element is the hamburger menu, and how it looks once you open it varies drastically from app to app. In many cases, it's hard to even tell which elements are active and which aren't (because everything is flat!), and if they're active, what exactly they will do if you try to interact with them. When I got her an iPad - the very first one - she really struggled figuring it out, because not only all the existing concepts didn't apply, but there was no rhyme or reason to it in general.

Sometimes the views you see here are extremely biased. It's normal, we use computers all day, we are experts. We learn and adapt.

For some people, a computer is just another tool they've been recently forced to use in order to live in society (in many countries, you cannot longer reallistically do your taxes without a computer for instance).

As programmers, and especially those who work in UI/UX design, we owe some respect to those people, because those who take the extra effort to learn a radically new technology at an elderly age, are completely, utterly confused when companies decide to move stuff around just for the sake of it (or as a result of A/B testing?).

And yes, sometimes the only way I've been able to teach people how to operate a computer is by literally describing the UI and the icons. In my experience, finding a good way to teach via fundamentals to someome who doesn't care is extremely difficult. And believe me that I've tried it many times.

OTOH, note that none of my points apply to early education (school). In that case, I completely agree we need to teach the fundamentals, not UI.

It's common occurrence having software dismissed as second rate because it's "ugly" or non-conforming to current design trends. Although form and function should enhance each other we've now reached a point where changes to form actively harm function. Toolbars are gone, menus are gone, icons are abstract outlines of what they represent, discoverability in GUI applications is hampered. I can find quite a few examples of applications both mobile and desktop that have become increasingly frustrating to use for no benefit whatsoever. "Where has this button gone? I'm sure it was here before"; "Oh I need to click the nondescript three-bullet button, then tap actions and there it is!"; "But the colours are now nice". Especially in a productivity tool I'd rather not have the interface change every other release because it's "ugly" by some abstract metric. Although I believe the ribbon turned out to be a decent design I'm fairly certain that wouldn't have been the case if Microsoft had changed again to something completely different within a release.

So, yeah... Please give me ugly but predictable interfaces.

While I somewhat get what you mean (and hate ribbbons), the Office UI is just way less crappy than LibreOffice's. It has smooth scrolling, some animated state, and they're at least trying to get rid of some visual and semantic cruft, so it's a lot less of a pain to use than LibreOffice.

Of course progress sometimes means breaking changes, software in general and especially LibreOffice is nowhere near a point at which we can afford to stop improving. It's not about changing things for the sake of changing things, it's about making things less painful (and hopefully ultimately delightful) to use.

I think this problem is not at all exclusive to propietary software. Take blender, for example. I spent quite a bit of time learning all the bits of the interface 5+ years ago. When I tried to go back, the software UI had changed so much that basically most of my knowledge and muscle memory is now worthless, having to google how to do every tiny operation. And this is not yhe first "huge overhaul" of the UI and keybindings they've made.

And it's not like you can easily get an old version of blender and make it work just because it's free software. Leaving aside the fact that I'd be missing out on newer features, system libraries get updated and at some point the old version stops working with a random crash. Then you need to start compiling from source and basically maintaining your own fork.

IMO this is what happens with large enough pieces of software where the devs care more about increasing adoption in spite of any happy existing users.

Sorry for the blender micro-rant, they are doing a great job. But still I felt it was the perfect example of microsoft-y behaviour in free software land

Meanwhile, I’m a prospective Blender user, and I hear it’s waaaaay better from 2.80, so that for various things where I might have used another program or given up and used nothing I would now be very likely to use Blender. It’s always a balancing act: satisfying existing users, against genuine improvement that will make it better for new users and existing users that put in the effort to learn the new way.

Blender is way better since 2.80 imho. I have been an a blender user since about 2005. I find 2.80 improved the UI considerably. For me it was well worth the little effort I had to put in.

I just don’t think the specific way to do each operation in the UI is an essential part of Blender. Anyone using it professionally or even as a serious hobby will pick up on UI changes quickly. The real work you do on Blender is the actual art, not just manipulating menus and buttons.

No offense intended, but isn't the real work you do in MS Office creating documents? Software is never about manipulating menus and buttons per-se.

It seems unrealistic to want new features but also be against any kind of refactoring or efforts to bring in new users. As you have found, it's time-consuming to backport things and the people who actually want that are not usually willing to do it. Probably because it doesn't bring in new users and therefore makes no money. Free software really doesn't change anything abut this, projects that want to become popular still need to prioritize growth.

Just use LaTex with Vim for documents if the org you work at allows it.

I find Emacs with AUCTeX to be the ideal TeX environment, but at least Vim is a sane environment. Word processors in general I find a very user-hostile environment for creation.

The ribbon interface it not just a "fad", it is a clear UI improvement since it removes a level of indirection.

If you wanted to see how a font looked in the classic Word UI, you highlighted a text and selected a font from the dropdown. When the dropdown was changed to show the actual font, it was the first step towards the ribbon, where the UI generally shows the actual effect of the options.

Programmers does not understand this difference, since they are used to work on abstraction levels removed from the actual output. So the ribbon seem like a fad just like the mouse and icons was seen as a fad by many developers back in the day. ("Why don't people just learn LaTeX?")

I disagree.

The number of times I had to use "Tell me what you want to do"-field is on a stupid level compared to the rich toolbar and extensive menus that existed pre-2007.

I mean ironically that's actually a much better solution - I love Jetbrains "search everywhere" field in Idea to find specific commands quickly.

So why in the heck the ribbon even needs to exist is baffling - just make a command search/command line input, and then if it's literally on a toolbar on screen, highlight it for me (configurably) so I as a new user know what to click on next time.

The ribbon is just somehow, always exactly wrong.

Great news. Now we have a search bar slapped into the middle of the title bar like an albatross. Nicely wedged in between all the other nonstandard title bar chrome.

I don't think the ribbon itself is a bad idea, but I absolutely hate the following two things:

1. Latency. When you click a dropdown, the whole interface can freeze for several seconds while Word presumably renders the menu or something. I can just about understand that for the fonts menu, but for the "bullets and enumerations" dropdown it's just plain bad.

2. Things moving around in the ribbon because Microsoft decided to push an auto-update to include some new search with bing or "data insights" feature.

I think they've improved it a lot since then, but I remember the Fonts menu getting incrementally slower as you piled more fonts into your system. At one time I think it actually held up Word loading entirely.

What’s so undesirable about teaching some technology in school and that technology changing over the years? Heck, I took computer hardware vocational classes in high school, and we learned about ISA, ATA, and PCMCIA. I would never bemoan either learning those things or those things changing. It’s not like that learning was made useless when the technology changed.

I think it is simply whether you are used to them only. My first time using Office is Offline 2007, the first Office with ribbons. When I want to use "old" UI in Office 2003 or LibreOffice(turning off ribbon), I also have problem finding the feature I want.

> I still have the same keyboard layout as back then but MS Office now has some janky "ribbon interface" which bears no resemblance to how it used to be.

This is a general pattern for GUIs as of late. It changes every so often until the point where if you haven't used the software in a while, you no longer know how to access a particular function via the GUI interface.

For example, web browsers used to have the standard drop down menus (file, edit, etc). If I want to view page source, the option would probably be under the view option and titled appropriately. These days, I have no idea where to find it, but at least I still know the keyboard shortcut to do the same thing and that hasn't changed at least.

Speaking of which, if you want a browser that still provides (opt-in) a classic menu bar with all the usual categories you'd expect, check out Vivaldi. Made by the guys who did the original Opera (before it went down the drain), so of course everything is configurable, stuff like vertical tabs is there out of the box etc.

The ribbon interface is one of the best things to happen to office.

It was horrible to use before that.

For an app as complex and feature rich as word or excel, the ribbon interface is a great way to organise commands.

Not to mention context specific menu tabs.

Especially in Powerpoint, my productivity has increased tremendously. I just create a table and automatically go to the last tab to format it.

> For an app as complex and feature rich as word or excel, the ribbon interface is a great way to organise commands.

With some reluctance, I agree with you. But it does bother me that I rarely use much of that complexity, while still being required to engage with a UI that seems mostly designed around hiding things that I don't need anyway. I wish there was a way to use the Office apps with a simplified feature set. Of course, I understand why not: everyone's essential features are different. So I carry on poking at ribbons looking for things.

That is a by-product, and a testament, to the complexity of the software.

Take Word. It has millions of users around the world. And everyone of them uses at most a different 20% set of the full features 80% of the time.

But, it's only one software. It has to satisfy everyone.

So you end up with a reasonably complex UI that everyone has to bear.

However, you can customize it, and I wish there was more documentation and more effort put into customization of the menus.

I mean, let there be a menu customization system that is fully drag and drop, rather than the clunky stuff presently there. It would be a game changer.

> The ribbon interface is one of the best things to happen to office. It was horrible to use before that.

Says who? Do you have any UI/UX study to back up that claim? Most of us on this page seem to believe the opposite is true.

> For an app as complex and feature rich as word or excel, the ribbon interface is a great way to organise commands.

Exactly the opposite. That is, a ribbon _may_ be relevant to a feature-poor application, but much less so to a feature-rich application.

> my productivity has increased tremendously. I just create a table and automatically go to the last tab to format it.

You're conflating context-sensitive UI changes with the use of ribbons. That's a different argument. Also remember that if a Table toolbar or sidebar appears, you can have the same effect - better perhaps - without ribbons.

For an app as complex and feature rich as word or excel, the ribbon interface is a great way to organise commands.

Not to mention context specific menu tabs.

Especially in Powerpoint, my productivity has increased tremendously. I just create a table and automatically go to the last tab to format it.

As mentioned elsewhere in this thread, Microsoft did extensive user testing in the course of development of the Ribbon UI:

https://news.ycombinator.com/item?id=23797590

> Says who? Do you have any UI/UX study to back up that claim? Most of us on this page seem to believe the opposite is true.

I never said people found it good. I said, implicitly, I found it good. Perhaps I should have made it explicit in my comment.

Also, it doesn't matter whether most of you believe the opposite was true. I was talking only about myself.

No need to be so, aggressive, perhaps.

> Exactly the opposite. That is, a ribbon _may_ be relevant to a feature-poor application, but much less so to a feature-rich application.

Compared to the alternatives, the Ribbon has more density, in the sense that it packs more buttons and controls than a normal tool bar.

Even AutoCAD uses a ribbon menu. I am guessing because the command density is much more than a normal toolbar.

> You're conflating context-sensitive UI changes with the use of ribbons. That's a different argument. Also remember that if a Table toolbar or sidebar appears, you can have the same effect - better perhaps - without ribbons.

My point was that the context sensitive UI is more apt to be used with a Ribbon system, than alternatives, like in 3DS MAX, where the context sensitive stuff comes up on a right-click. Or in Office itself after we highlight an item.

I reiterate, it is a personal opinion. No need to get so defensive about it.

Some people want their commands and work area to be separated. Some like it otherwise. I prefer the former. When I click my table in PP, I don't want something overlapping it with a menu item, that is not stable, and disappears after a while or after another interaction. It's distracting to me.

> Consistency is worth so much more than fad-chasing. Learning something is more worthwhile if you are able to trust that it won't change underneath you.

Almost anything could be described as a fad. Interactive computers were seen as a fad once. Punched cards were good enough! The command line was good enough!

For complex applications (roughly defined as, lots of nested menus), the ribbon UI is an improvement as it surfaces more features. In itself it's an evolution of the customized toolbar approach - there's plenty of UX research that showed that most people only had the standard toolbars. (Office still does offer a context sensitive "toolbar").

> we had an exam where we got a printed out page and had to reproduce it in an hour. This was supposed to be fundamental computer knowledge like learning to type.

I'm hoping this didn't mandate a particular program, but even assuming it did -- Microsoft Word's basic shortcuts all still work exactly like in 1995. The common toolbar buttons (bold, italic, lists, etc) are in the same place. Even the original keyboard shortcuts for deeply nested menus (e.g. Alt D F F, in Excel, for showing the filter UI, i.e., the old 'autofilter' command) still work.

So where is the problem exactly?

>Almost anything could be described as a fad. //

I think the key thing here is that MS Word had the features most people need back in Win3 days (possibly earlier?).

There's stuff you want to do that punch cards don't allow, eg duplicate a program virtually; but there's not stuff people wanted to do that menu based wordprocessor UI prevented.

Ribbons fine, but no advance for me over menus, on balance I prefer menus but that might just be familiarity and age.

> I think the key thing here is that MS Word had the features most people need back in Win3 days (possibly earlier?).

That's a very narrow view of what people need. Children at home these days "type" essays into Word by dictating as a first-party feature. This is a pretty recent step forward -- in the past it used to be a very expensive third-party add-on with much less accuracy.

Good, scalable multi-user document editing has changed the way students and even workers take notes in classrooms and workshops. Again, a fairly recent phenomenon (although Word still isn't as good as Google Docs here -- c'mon Microsoft, keep up).

My kids are recent high school graduates. When they were in grade school, they learned MS Office. They easily picked up the newer stuff as it has come along, and at the same time, the schools have gotten a lot more flexible. It just wasn't a problem. Today they use Google Docs or whatever.

At my workplace, I see gradually declining use of any kind of word processing. Folks just edit in the e-mail or chat editor. One colleague makes video blogs.

Word processing seems to remain, largely for documents that will never be read, such as HR announcements and dissertations. ;-)

I've written lengthy documents in the chat editor, and have discovered to my great surprise and pleasure, that people actually read them. One thing that seems to be important is for people to be able to read things easily on their phones.

MSFT had reason to believe that the Ribbon was more accessible and discoverable to users than the old menu structure. I was dubious at the time, but since the shift I have come to believe that they are correct. I think the Ribbon is better for the vast majority of users.

Interfaces shouldn't be changed for no good reason -- and MSFT is certainly guilty of chasing fads or tweaks without a clear benefit, too, in other areas -- but neither should an interface be frozen in amber if a better idea comes along.

(Also remember that mass market software isn't designed for people who post on HN.)

> MSFT had reason to believe that the Ribbon was more accessible and discoverable to users than the old menu structure. I was dubious at the time, but since the shift I have come to believe that they are correct. I think the Ribbon is better for the vast majority of users.

Microsoft also removed the start menu on Windows once and everybody hated it. Sometimes, MSFT is wrong.

I would absolutely never dispute this, but I also don't think it has anything to do with my statement.

Next thing you know they'll start insist on adding lowercase characters to the character set and forcing us to use color displays and Arabic numerals instead of good old fashioned Roman ones!

I believe schools are mostly wrong in their computer education approach. The purpose and functions of MS Word (and most word processors for that matter) have not changed significantly in the last couple decades, despite the change of paint. Schools should be focused on teaching people what common applications are for and what they are capable of - not which buttons to press to perform specific tasks.

> It's weird to me that people complain over "old and ugly" interfaces.

You might be interested in checking out the work of Donald Norman. He makes a great case that attractive things work better. Old isn't necessarily a problem, but ugly is.

https://jnd.org/emotion_design_attractive_things_work_better...

This seems intuitive enough to me but years on HN have let me know it's a rare opinion, at least among developers.

I 99% agree with you, except that I believe LibreOffice is not just old, it's subtly bad. I didn't notice it before but it's too adhoc, just too laggy, just too crummy at times. It lack tastes, consistency and responsiveness that you got even in office 2003.

I wish LO to improve everything, they came from far and did a lot. But it's not enough.. kinda like blender 2.7 to 2.8.

Isn't Blender 2.7 to 2.8 widely considered a successful redesign? Blender is experiencing continuing and dramatic industry support. I've heard Blender is comparable to the proprietary alternatives.

I've not used Blender much though.

Oh sorry, bad phrasing, I meant right now LO is stuck at blender 2.7 and needs a 2.8-like push.

It's not just slow, but just the subtle things in how it behaves. For instance, try navigating a spreadsheet in LO with only the keyboard. It's super annoying. Then try the same thing in, say, Google Sheets, and it does exactly what you think it should do. And that is a huge difference in terms of UX and productivity.

yes, UX is stuck in 90s way but not the best part of it.

I wonder what's the state of the codebase and if it could handle a significant UX redesign.

I wonder what people who have spent their adult life studying and implementing UI/UX think about it.

If there is one thing I would suggest for all large open source projects, it's separating the functionality, data model etc. from UI as much as possible.

It should be possible to have two separate open source projects. One is just extensive framework library with good API and the another is user interface. This would allow multiple separate UI's, web interface, command line, mobile, tablet, etc.

> Consistency is worth so much more than fad-chasing.

Consider consistency across applications though. People like conformity.

A good interface should not need to be learned. Everyday new people are born that have never used a bad interface before. There is no justification for keeping bad interfaces around just because you can learn them.

Interfaces that are hard to learn are also hard to recall. These lead to mistakes.

The only thing I want LibreOffice's UI to do is get the fuck out of my way.

I also took a pre-ribbon exam for Word in high school, and I’ve had no trouble applying those concepts to successive versions, even though I’ve never been a heavy user. I dare say most software doesn’t change quickly enough for intensive early use to become totally irrelevant.

That wasn't to make you memorise a GUI but to make you understand the connections between visual indication and concealed action.

Without "fad chasing" (which is a sunk cost, a week every few years of pro graphics design), without even copying microsoft office's layout, you are losing out on new users. Why would anyone use libre office because it's open source rather than being a better functioning product that is more pleasurable to use (and look at)?

> Learning something is more worthwhile if you are able to trust that it won't change underneath you.

I think there is something even deeper than this. The usability of UI is directly proportional to quality of users' mental model of what the application's doing.

Minimalism can be helpful for removing the "magic." I like the way Google Docs has evolved their UI. They started matching the feature set of desktop applications. Slowly, they have removed and simplified features, leading to a beautifully minimal yet powerful interface for typesetting and writing documents.

While Microsoft has changed a lot of stuff, the keyboard commands in Windows and MS Office have remained mostly consistent in my experience, so when trying to work on someone else's computer, some things I could still figure out easily enough even though the system seems different. This could be done just as well in free software projects too (some of which already do, I think).

(However, for my own stuff, and on my own computer, I use neither MS Office nor LibreOffice nor other similar software, since I do not need them; I can use plain text, I can use TeX, I can use PostScript code (with Ghostscript), and whatever.)

In high school they taught us Lotus 123 spreadsheets, oh my do I feel old! Never used those again afterwards. I agree about the proprietary software gripe, but it could have been worse.

This "once you learn the complicated non intuitive user interface it becomes good" mentality is the reason why a lot of people don't like Linux (in my opinion). The whole Linux world just has such a poor UI/UX design that you don't want to interact with it in the first place - not all but most of it anyways.

There seems to be a very large elephant missing from that discussion which is the rise of SaaS office suites. Speaking personally, I'm very glad that there is a free software office suite (as well as drawing programs etc.) However, I sincerely hope that collaborating on documents and presentations by mailing them around to people and trying to merge their edits and comments is never again the norm for me. Shared documents are probably a much overlooked enabler for people to work remotely in the current situation.

Added: They do mention LibreOffice Online but that seems like it should be a more central point in the discussion.

I agree, the main point I would expect from "The next five years of LibreOffice" would be a self-hosted competitor to Google Docs and Office 365. There's definitely a place for a local application editing documents, but I don't think that's what's going to matter the most five years from now when compared to a solid web based alternative -- and I also don't think a "code freeze" for the "normal" LibreOffice (not that I'm advocating one) would make the product unusable five years down the road.

It's there but with the clear and explicit intention that the free version will lag behind the commercial offers.

They have a self-hosted online suite, it's called Collabora.

It's a spinoff

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

The former LibreOffice development team from SUSE joined Collabora in September 2013 forming the subsidiary Collabora Productivity.

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

I find it really odd that they chose server side rendering. I only did a very brief demo of it, but I was not impressed.

Yes, unfortunately. Makes impossible to work under unstable connections.

I hope Webasm enable porting at least some of the rendering code to client-side in the future.

Why not just use a desktop program with online features, then?

I'd assume that's less of a "chose" and more of technical limitation caused by wanting to reuse all the legacy code.

There's definitely a market for self-hosting as in the case of Nextcloud, for example. There are lots of reasons companies, much less governments, may not feel comfortable using Google or Microsoft SaaS offerings.

But, as you say, that's different from running a local application. While some heavy-duty multimedia still benefits from running an app locally, that ship has mostly sailed with office suite software. I still have LibreOffice installed on most of my systems, but even if I'm not collaborating on a document, it's incredibly useful to have docs that are "just there" and I can access from any device--even including my phone in a pinch.

I don't think making a web-based version would make sense for TDF. It would basically have to be done from scratch and OnlyOffice is already years ahead of them in that. Any attempt to port the existing code to the web also seems destined to fail (see Collabora).

What I think would really make LO stand out is strong collaboration support in the existing desktop editors. A self-hostable headless sync backend that streams real-time changes from and to everyone who has the document open would be even better than GDocs, which is notoriously slow due to being all in a browser.

Collabora is an open-source, self-hosted, online collaborative office suite built on top of LibreOffice. Take a look, you may find it useful.

Agreed - students have already moved to google docs or MS Word with OneDrive. We are used to giving feedback using these tools, students making changes and getting those signed off by us.

Full history of changes. Full history of lecturer comments. Backed up online. Editable on phones, laptops, PCs...

This is the world they are expecting when they enter employment, or start their own start up.

To be honest, I'm a bit surprised that there's still the level of apparent interest in standalone LibreOffice that there is. I work somewhere that was almost exclusively a LibreOffice shop but we adopted one of the online services. Although we didn't really force people to switch, I don't know the last time I've seen a LibreOffice document even if I imagine some people still use it.

Office suites are a huge piece of the pie so even minority use cases get a decent number of users.

> They do mention LibreOffice Online

And it says there is revenue there, correct? Maybe that's why it isn't expanded.

It's been a few years but I had a client that needed some secure remote file access capability "like DropBox" so I spun up an ownCloud VM and it had a surprising amount of collaboration tools built-in. It had all the groups, permissions, sharing, etc stuff you'd expect, and you could open and edit Office documents in the browser (it was a web-based Zoho Word/Excel type editor if I remember correctly.) It also had automatic versioning so if someone jacked up one of the files, they could roll back to whenever.

For anyone looking for that sort of thing, ownCloud might not seem like an obvious choice but I was pleasantly surprised.

IMO if LibreOffice really want to be in it for the money, they need to move away from trying to catch up to Office (not possible) and target some other niche which actually differentiates themselves. The fact that Airtable and Notion are around and pretty successful means that people can and are spending money on document management solutions that are not Office.

A better API is one such point. Microsoft APIs always suck, and LibreOffice could be the thing that has a rich interface that allows you to build XLSX and DOCX files via an API. For a product that has been around for more than 20 years (if you count StarOffice), I expect the API story to be much better than https://api.libreoffice.org/examples/examples.html

Democrats in the US House of Representatives just approved a $741 billion defense spending package. It will almost certainly clear the senate and be signed by the president.

Imagine taking enough of that money to support these 40 LibreOffice developers and maybe say 20 more. There would be no need for these acrobatics around source availability.

I understand why the companies need to do this. But there is an existing system for funding things with widespread benefits to society. It’s called the government.

It begs an interesting question: does defense spending have a positive ROI? I don't know the answer to this question. But I'm thinking funding R&D, trade route protections, and perhaps some nefarious things can all contribute positively to American interests and thus American society.

Ginormous defence spending is how the US do a "social state". That's pretty obvious once you see how defence money is peppered all across the country.

It's a republican jobs program.

Nah the defense industry is smarter than that - it's a bipartisan industry. F-35s have components that are built in every state for a reason.

It doesn't mean its productive. We could be spending half a trillion on civil engineering to rebuild cities, roads, high speed rail, canals, etc. That has appreciable and visible benefits to society.

Occupying a desert on the other side of the world largely to disrupt oil pipelines seems less effective as stimulus.

Which also begs the question: How many such projects in different industries (eg, agriculture, aviation, food, healthcare, etc) would need a slice of that defense pie?

---

[Note: I'm not advocating for/against the idea of the parent comment, just adding a layer to it]

> "It begs an interesting question: does defense spending have a positive ROI?"

It's analogous to the question: Does homeowner's fire insurance have a positive ROI? The answer is no only if the homeowner's house doesn't burn down and it's enlightening to look up how many house fires occur daily.

If that's not compelling then perhaps contemplate what would people have said the ROI for national readiness for a pandemic was a decade ago and what they would say today.

>Imagine taking enough of that money to support these 40 LibreOffice developers and maybe say 20 more.

There are infinite ways to spend every dollar, why is LibreOffice funding worthy? In other words, why is LibreOffice's popularity so low that it cannot find a way to fund its operations? Which then raises the question - why do you want to fund an unpopular project?

I think the idea is that 1% of a b52 bomber might take LibreOffice to an unparalleled place for a libre project.

But why is LibreOffice so special? Surely there are thousands of other projects that would also love 1% of the b52 bomber budget.

I get the 'idea'. There are more things to fund then there are funding dollars available ... by many orders of magnitude, so you have to make decisions. One of those decisions is whether or not to pay for a certain numbers of b52 bombers. Another, completely separate decision, is whether to fund a very specific niche open source project. Linking those two is disingenuous. If you want funding for LibreOffice, make the case to your representative. Make the case to others so they make the case to your representative.

Sure, but it could also take other projects to unparalleled places. Why does LibreOffice deserve that money over other projects?

As schools move more and more online they are locking into proprietary office programs that suck funds from schools, and students can't afford. Sure, as long as they are students they can use the school's software for free, but once they graduate they can't.

So in effect the government is already funding google docs and office 360. I would prefer the government fund opensource software that anyone who pays taxes can use for free.

>So in effect the government is already funding google docs and office 360.

Government buying services, or software, or anything, isn't really like 'funding', it's paying for services rendered.

>As schools move more and more online they are locking into proprietary office programs that suck funds from schools

And who is to say that LibreOffice is a good alternative. I'll tell you right now, it's not - which why it isn't really being used. It's a pain in the butt to deploy, and support at scale (from identity management, to sharing and collaborating, to security fixes and patches).

>I would prefer the government fund opensource software that anyone who pays taxes can use for free.

Sure. I have a preference for a lot of things too but we live in a democracy, so nobody gets to dictate their preferences. But feel free to engage with your local school board and make your case. Feel free convince your peers to make the case with you.

While not open source, you can definitely use Google docs for free student or not.

But it is a proprietary, private owned software. That can change anytime. Maybe not in the next 10 years, but it can and I suspect that it will.

Maybe, but

>So in effect the government is already funding google docs

Is just purely incorrect currently.

I think the cost should be compared to whatever they pay Microsoft for Office subscriptions for the federal government.

It's not really worthy of funding anymore. Certainly not by the government. It's just not competitive with other options. I used to be a longtime Libre Office user. Back then, MS Office was expensive and only available on Windows. Even the Mac version wasn't completely compatible.

MS Office is better than it's ever been, with a variety of licensing options that are affordable. And if you're a student you can get it for peanuts or free. Being able to use the same tools on desktop, web and mobile with everything accessible in OneDrive is a no brainer and outstanding value for the high end price of $99/yr for a family plan. If you're on a less expensive individual plan, a student plan or a discounted plan through your employer's HUP benefits, it's really kind of an easy decision.

As a counterpoint, I imagine the amount of money the US government already spends on MS Office could fund development of something as good or better than Office, while giving it away free to the rest of society improving productivity. Although, I wouldn't bet on the management of that project playing out like that.

>I imagine the amount of money the US government already spends on MS Office could fund development of something as good or better than Office

No, it can't. Microsoft lives and breaths Office 365 because that is a core part of their business. There is zero chance that a government will care as much about this as Microsoft does, or the 10,000 other companies that are trying to supplant incumbents. Add to that the fact that funding will be uneven, because if the tool achieves a 'good enough' status, why keep funding it when you have all those other interest groups vying for money. The project will be susceptible to lobby and interest groups. Maybe a Senator will see it as a job-works program and move development to Alaska to provide Alaskan programmers jobs.

Time and time again these massive government-run software project collapse on their face.

In the commercial space, government just cannot compete with private industry ... at all. And this is not a libertarian disparagement of government. Government has a proper function, and competing in software with commercial offerings is not one of them.

>while giving it away free to the rest of society improving productivity

Says who? In fact, this is guaranteed to not be the case, because who is to say that funding a word processor is a productive application of capital when you're not responding to market forces.

> Time and time again these massive government-run software project collapse on their face.

Which is why I said "I wouldn't bet on the management of that project playing out like that." I also completely agree Microsoft is way too entrenched to start something today.

> In the commercial space, government just cannot compete with private industry ... at all.

I kind of think you're overstating what I was suggesting (likely, I stated it poorly). Instead of paying Microsoft $x annually for MS Office, I think for less money they could write and maintain something that fits their own needs. Since it was developed with tax dollars, release it publicly. I doubt businesses would adopt it unless some independent companies popped up to facilitate it (like Red Hat does for OSS).

The goal wasn't to replace Microsoft Office in society, just to fulfill its own needs.

Yes, it can.

We have federally-funded national Research labs that regularly take on nuanced, technical challenges like this, and they work great. It's not like a federal bureaucracy designed and built multiple mars rovers, engineers did.

>Add to that the fact that funding will be uneven, because if the tool achieves a 'good enough' status, why keep funding it

If it is good enough, sure, it doesn't need the same amount of funding required to get it to that point. That is a good thing.

The federal government isn't the only potential source of government funding.

>In fact, this is guaranteed to not be the case, because who is to say that funding a word processor is a productive application of capital when you're not responding to market forces.

I don't follow. Because we can spend less money building a thing once than renting it annually for decades and decades, which is a response to market forces.

>We have federally-funded national Research labs that regularly take on nuanced, technical challenges like this, and they work great.

Not in commercial space. Government is good for funding cutting edge research with no obvious commercial benefit. That's a good role for government. Government is atrocious at competing with commercial offerings.

>It's not like a federal bureaucracy designed and built multiple mars rovers, engineers did.

Right now there is no way for private industry to do better than NASA because there is no clear commercial benefit and the money to develop potential commercial benefits are completely prohibitive. The missions that NASA engages in, from rovers on Mars, to New Horizons probe, to James Webb telescope are very important, and could not be done by private industry (though private industry is sub-contracted for many parts of those projects). That's a proper role for Government - identify gaps that the market has left but that have societal importance and fill then with funding from taxpayers.

>If it is good enough, sure, it doesn't need the same amount of funding required to get it to that point.

Word processing has been 'good enough' since 1995 - why isn't everyone using Office 95, or LibreOffice.

But you do demonstrate why government software projects fail - there is never and 'endpoint' (or 'good enough' point) to software. You're always updating based on market needs.

>Because we can spend less money building a thing once than renting it annually for decades and decades

I reject the assumption, it won't be cheaper to move off of Office. It won't be cheaper to build an alternative. Again, why isn't LibreOffice everywhere? The product is free. Why are companies paying for Office? Are they stupid? Maybe because there are hidden costs, from maintenance, deployment, collaboration with other sites, training, etc. that make using LibreOffice actually more expensive in practice than Office.

And by the way, for many companies, Google Suite has supplanted Office (which is typically a paid product). But again, somehow the fact that LibreOffice free, didn't translate to being competitive with that offering.

This is exactly what Sun thought. They looked at their licensing bill for using MS Office and thought they could develop an Office suite for less than that themselves. They bought this little German company which made "Star Office" and started developing it for their own use and also selling it. Turns out, it didn't work. So they Open Sourced it, called it OO.org... it still didn't work. 20 years, many rebrands and many changes of ownership later, that same piece is software is still struggling to find a way to exist. Turns out, developing a competitor to MS Office is not that easy or cheap.

The government writing software for their own use is quite a bit different than Sun buying or writing one without customers lined up.

Do you think it would be right for the government to fund a competitor to an American business? Wouldn't it make sense for the government to just use or license Libre Office if it was good enough to meet their needs?

Eh, I don't see the problem. If the government decided not to renew a contract with SpaceX because they thought they could do it better themselves, should they not? Tax money spent on government research should be public. I do think they should follow any of Microsoft/SpaceX's IP laws (but also file their own for the public).

> Wouldn't it make sense for the government to just use or license Libre Office if it was good enough to meet their needs?

Sure, if the math and logistics worked out. I kind of think Office and Microsoft are too entrenched now.

I think the reason it wouldn't be a huge deal is you would still need some sort of business arm for it to get traction--like a Red Hat. So basically Microsoft would just lose a government contract.

Imagine paying a subscription to access my own documents! Paying for privacy to be hoovered up to another corporation.

Sure, I bet they have some good features that I probably wont use.

> But there is an existing system for funding things with widespread benefits to society. It’s called the government.

Not if you ask the current ruling party. Even a cursory glance at the state of the Union should make it obvious there's nobody left who cares to fund things with widespread benefits to society.

A great example of this working for real is Ghidra[0]! Funded by the government, a spectacular piece of open source software.

[0] https://ghidra-sre.org/

The funding necessary to support LO development is so tiny relative to the US defense (read: offense) budget, that it's not even worthwhile to make this reference. The US DOD's stationary budget would probably cover this.

As much as I love free software, LibreOffice interface feels old and ugly (to me, at least). It would be nice if there was more focus on the UX/UI design in open source projects. Of course, most programmers are not UX/UI designers, and designers don't tend to work for free on open source projects.

There's also the question of how open open source programmers might be to some UX/UI designer stepping in and trying to force their vision on the project.

Beauty/ugliness is subjective. For some it is "old and ugly", for some it is "retro".

I agree, and programmers (at least on HN) seems to prefer minimalist and/or retro interfaces.

However I'm pretty confident that if I show such an interface to non technical people, they will feel that it doesn't look nice, compared to whatever proprietary app they use on their phone, Windows/Mac, ...

I personally don't use Word nor LibreOffice (I use mostly LaTeX), but if we want free software to become more popular, there must be a focus on the design.

I don't care how ugly it is. I care about how quickly people can get their work done. The ribbon is better for this because the things most often used are right there and easy to find. Sure rarely used stuff takes longer, but that is a good trade off.

Is there some way to move stuff in word so that it is more accessible. Just wondering.

Yes, since ages ago the ribbon is customizable. You can create a new tab with exactly what you want in it.

There is also a search feature now, so its not like you need to know where rarely-used things are.

Well, the head of the design team is Heiko Tietze. His PhD is in Psychology and he's worked at various companies as UX designer/lead.

So, why you'd you contact him and tell him where he's going wrong? I'm sure he'd value your insight and guidance.

I don't personally mind people putting in a fancy ui, put whatever ui you want on it.

Just also give me keybindings and let me hide whatever ui you make.

On the other hand, professional UX/UI office software designers came up with the ribbon.

No... the problem with desktop software is that creating a good UI/UX is a hard software problem typically solved by software developers instead of designers. Whereas, the accessibility and power of web technologies show that making UI/UX an easier software problem (HTML, CSS, JS) automatically attracts more designers yielding better UX/UI.

Your first point is correct, but as a ux designer (not what I do, but it is in my background), most web ui is terrible. It is done by designers, but they obviously have a background in art not user machine interaction and it shows. Once the page loads (takes too long Turing off users) it looks nice, but try getting anything done.

>making UI/UX an easier software problem (HTML, CSS, JS) automatically attracts more designers yielding better UX/UI

I find this a bit questionable. What are some examples of this good UX and UI? Many HTML-ish UIs I can think of are not good. Slack, Discord, modern Windows apps such as the new settings app for example. They tend to suffer from poor reimplementations of native widget functionality, poor implementations of scrollable elements especially infinitely scrollable ones, low information density and too much whitespace often exacerbated by flat design, janky responsiveness often with high latency.

Apart from the remarks about flat design, whitespace and information-density, which are very personal imho, I would say Google Drive, Google Docs, Microsoft Office Online, GitHub source code editor, Visual Studio code, Atom, all do a good job.

Good job? Maybe on a beefy first-world computer. The UX goes down the drain if you try to run a few of those apps at once.

Doubtlessly the programmers who work on those products use computers with dozens of gigabytes of ram and half a dozen cores or more. They're out of touch with the sort of computers a TON of casual computer owners are using daily.

Github Desktop I'd say

The products you mention are all hugely successful, often against the backdrop of existing OSS with similar features (IRC, for example).

Maybe there's some abstract notion of "good UI" that is distinct from "people like to use it". Or maybe people are stupid.

But on a purely empirical basis there is absolutely no evidence for the supremacy of the old UI paradigms you champion.

Many of the things I'm talking about aren't even debatable they are simply bugs. For example infinite scrolling areas jumping around as they load content and the interface having variable and high latency in response to user input. 'Poor reimplementations' could be stuff like copy and paste being bugged out in a text box.

In any case you're right, I think there isn't enough study into these things.

> the accessibility and power of web technologies show that making UI/UX an easier software problem (HTML, CSS, JS) automatically attracts more designers yielding better UX/UI

"Modern" UIs are often atrocious, and I'm not the only person who thinks so.

"better UX" is not a thing I [This comment uses cookies to enhance the user experience. Click :HERE: to accept] would ever say about websites over desktop software.

Whoa.

I much prefer finding my way around any "badly designed" UI that can respond in 20-100ms a "well designed" UI that responds in 500-5000ms.

Most web software providers don't (can't!) deal with server latency or the client-side processing of the JS/CSS etc. that they use. That's why I use a lot of "well designed" UIs that respond slower than equivalent systems from the 1980s.

UI/UX isn't a software problem at all. It's a design problem, and designers solve it no matter the platform.

I find it reasonably straightforward to use. I can find most things easily in the submenus.

Though I'm on the side of classic menus in the infamous ribbon dispute.

LibreOffice interface feels familiar (to me, at least), and I fell productive with it. My only complain is that the application itself is quite heavy and takes quite some time to start in my Macboook (late 2013)..

LibreOffice has a ribbon!

Yes, unlike Microsoft they offer you a choice of interface. I'm sure if MS had done the same there wouldn't be nearly as many complaints.

Thank you so much for saying this. I had no idea until I read your comment and immediately opened the view menu. I am blown away I didn’t know about this after over 10 years of using LO...

It's still pretty ugly and doesn't work as well as the office version, though.

(LO Dev) We actually have quite an active ui-design sub-group, which tries it's best to refresh the look and feel.

Feel free to join in :-)

I hate to rant, but there is a weird sense of entitlement in many comments. An assumption that software should "just work", be completely bug free, have the best design, have developers quickly responding to bug reports, etc, while giving away the product for free.

I might not use an office suite very often, but when I do, I'm grateful it exists and is free/libre. Some orgs (gov, companies, service providers/resellers) do rely on LibreOffice, and they should really make sure that core developers can keep on going.

(I work for a free software service provider/reseller and core developer, this comes up regularly)

This is the de facto complaint of open source software, and initially I thought the entitlement can really only be explained by laziness, after all why don't people just bother to read the source code and contribute the feature they want?

But then I realized its way way more rudimentary than that. A substantial amount of software developers struggle w/ interpersonal communication skills, and so often what may have been intended as a polite feature suggestion came across as entitlement, if not bewilderment that a feature doesn't exist through a poor choice of words combined with a lack of empathy that we see everywhere throughout the internet anyway.

> why don't people just bother to read the source code and contribute the feature they want?

It is a bit far fetched to expect non programmer users to read the source code, much less contribute the feature (i.e. from requirements gathering to feasibility analysis, to design, implementation, testing, and release) isn't it?

That is the kind of work that only a compensated team would do, especially and something as complex as an office productivity suite. In other nonprofit areas, your organizations doing work rely on charitable contributions from donors who share the same interests. Until such a time that such donors exist, it's hard to see how the development model is funded.

> A substantial amount of software developers struggle w/ interpersonal communication skills, and so often what may have been intended as a polite feature suggestion came across as entitlement

They might have interpersonal issues, but it seems like the deeper issue here is more with the incentive model, not with the developers' personality traits.

So you say most of the users can't utilize freedom to change program. Neither coding, not testing, not money, not individually, not collectively.

In other words - they take someones else work for free and give nothing back. And then they complain about quality.

To be fair there is simple free open source software that just works, like vim. It is just that people should understand that with complicated and necessarily bloated software which does a lot of things major bugs are inevitable basically fate.

In the early days, one of the things that was heavily hyped by FOSS advocates to drive adoption was the "free as in beer" aspect. It worked but, unsurprisingly, it resulted in a large userbase that only cared that the free beer kept flowing and nothing else about the FOSS philosophy nor contributing.

On the other hand, the idea of FOSS is to put effort of a few for the benefit of them and many more others, a positive sum game.

In some cases this idea is able to deliver, in others it isn't. But the expectations of having a working, maintained and free product are justified to a degree.

I think there is a quasi-unique problem with Libreoffice. People love Word and Excel, in ways they don't Windows and other pieces of software that compete with FOSS.

Those two programs are so well polished, user-friendly and powerful, and are what so many people are trained on, that it can be rough transition to LibreOffice.

Well, I believe that FOSS software should "just work", but also that we should "just support it" if we use it - with decent bug reporting always, with QA work if we can spare a bit of time, with coding if we know how and can spare more time, and with some money if we're not poor.

So, I guess I believe in "mutual entitlement".

Why so much talk about the ribbon? This article has nothing to say about that. This is about a Personal Edition and an Enterprise edition and how little revenue its actually coming in.

Basically, libre office won't be libre anymore. Looks like Enterprise is gonna have a commercial license.

By the way, I think the ribbon is far better than the ugly dropdown menus. It's also been over a decade since it was introduced. Some of yall are so resistant to change is bordering on zealotry. I hate to remind everyone about dropbox but dropbox.

> Basically, libre office won't be libre anymore

The personal version still will be, right?

Now this would be something to worry about if all new feature development were put into enterprise, and the personal version stagnated.

But from my experience in enterprise software, enterprise features are generally not even applicable to consumer users at all. They're things like single sign-on, approval workflow integration, private cloud integration and whatnot.

It's not as if a bugfix for page margins will make it to the enterprise version and not the personal one.

So this doesn't feel like something to worry about. If it provides the necessary financial resources to keep the personal edition alive, then I'm all for it.

Yeah I’m cool with it too, it’s just not quote unquote true free software anymore. Which should be big news, but basically the whole first page of comments is all about the MS word ribbon

Madness of crowds for sure

It might sound stupid, but looking at the marketing plan attached to this article ( https://nextcloud.documentfoundation.org/s/jzryGw7XDkJadmo#p... ) I wonder how they could deliver such an important message in such a messy way. There are context slides together with strategy slides, together with marketing slides, together with analysis slides. the best one is slide 48, which seems to be a leftover of the initial storyline?

I really like LibreOffice and I try to push its usage where I can, but reading about a potentially risky development strategy in such a messy way makes me wonder if it has been thought out so well or it is a muddy idea reflected in unclear writing.

Huh. I just tried Calc (6.0.7.3, Ubuntu 18.04) again this morning for the first time in a while and it crashed in seconds after right-clicking a few column headers. It's repeatable.

When trying to navigate the very confusing document recovery dialogs to restart it, it then launched as Writer not Calc.

I really wish it was an option but JFC. My experience with has always been so bad it doesn't even feel worth raising a bug report.

I find that calc is very good for migrating data from files that other programs, cannnot open. Import from some random file type, export as csv for example.

I've found that Calc is better for importing things like custom format CSV files etc. But has problems opening very large files, where Excel is better able to handle working with them.

Overall that makes Calc more useful to me, but then again, I mostly write scripts when dealing with data files if I can.

Could you make a video of that? I am clicking myself silly (ubuntu 18.04 / 6.0.7.3) and cannot repeat that behavior. I use calc a lot and it rarely crashes (less than excel/windows which, on my ms surface tablet, actually quite regularly crashes or hangs).

Sure, here you go:

https://imgur.com/a/PZMaqff

org.gnome.Shell.desktop[4355]: Window manager warning: Invalid WM_TRANSIENT_FOR window 0x9000024 specified for 0x9000965 (LibreOffic).

You have shamed me into raising a bug ;)

edit: raised, the 6.0 branch is EOL, doesn't happen in 6.4.5.2

Ah, I am using i3wm as wm; that might be the difference...

I am happy to report it doesn't happen with 6.4.5.2.

It's a shame that the snaps installed by default in Ubuntu haven't been updated.

yup, that's the thing, people love to complain on random social media platforms, but most people can't be bothered to report bugs. been guilty of it myself as well of course

The sad thing is that it's almost futile raising a bug against complex software like this or IDEs. It doesn't matter if it's FOSS or commercial, they already have backlogs of 10000s of bugs so unless loads of other people are reporting the same bug, it almost certainly won't be fixed.

I have never had a Visual Studio or Windows bug report fixed and only a few JetBrains ones fixed. However, if it's a small tool or library then it's a different story, much higher response rate.

> It doesn't matter if it's FOSS or commercial

It does matter, you can't fix a problem with IntelliJ, you need to wait until the company considers it economically viable to fix, ie enough people report it or its critical.

But if something in an Open-source system pisses you off, you can fix it yourself.

It not for everyone, but at least it's possible to fix FOSS systems yourself...

Oh sure, I was just meaning in terms of bug reports actually precipitating action. For this type of software it's more like software a complexity issue than an economic one.

> But if something in an Open-source system pisses you off, you can fix it yourself.

I'm literally doing that right now :)

As someone who frequently encounters bugs, let me explain why I don't usually bother to file bug reports: No one cares.

Choose any large long-lived open source project at random and search their issue tracker and I guarantee you can find at least 3 open bugs that are 5 years old or more. I know, because I've run into these bugs, as have dozens of others, and went to report them only to find they had already been reported years ago with no action taken by the developers.

Filing bugs takes time. You're going to want reproducible steps, you're going to want logs, you're going to want me to try different things. That's fine, and I'm willing to do that to get my bug fixed, but only if I think you actually give enough of a shit to fix it. I don't enjoy wasting my time any more than you do. When I see 5 year old open issues your tracker, where several users have chimed in, submitted logs, etc. and you've done nothing, I assume you don't care so neither do I.

Worse yet, many developers are downright hostile to users reporting bugs or, god forbid, making feature suggestions. Spend five minutes in GNOME's issue tracker to see what I mean.

For example, I encounter this bug every day when I open Firefox (latest version) on i3wm. "Opened 9 years ago."

https://bugzilla.mozilla.org/show_bug.cgi?id=686747

Bug is open, "firefox57 - wontfix" means "would not be fixed in Firefox 57"

* Create New Account

* Edit Bug

* Tracking Flags: $version => affected

* Save Changes

Do it for every version. It is extremely hard to know is issue resolved or not. I've tried to reproduce session restore on Developers Edition - nope, I can't. There is no minimal xmonad config attached. Is it solved by ewmh (helps at least with fullscreen)? If bug "is simple to fix" try to fix yourself.

For a sufficiently large project (let's say 100k issues+), I think it's totally reasonable for some vanishingly small percent (say 0.003%) of issues to be really quite hard to reproduce/debug/fix without breaking other things. I don't think that in any way implies the maintainers don't care, and to not submit reports because you fear you'll end up in the 0.003% rather than the 99.997% is a bit silly if you ask me.

I maintain such a project, and most bug reports I can fix in a few minutes. Others take a few days. Still others need to be deferred until some large refactoring is completed in coming months. Others still are involved to the extent that I'm not sure I'll ever fix them. But I value them all, and there are many more in the former categories than the latter.

Yes, there are hostile developers. Yep, even pool requests with unwelcome features may be rejected (we prize them as focus in proprietary software).

Other... I can turn your rhetoric around: "Fixing bugs takes time. There is source code, yet in 5 years no one fixed. Looks like niche issue. You do not care about it so neither do developers."

Hey, it is not bad, at least you've confirmed you are not the only one. One can't even browse Opera issue tracker.

So, I'm on the same version of Ubuntu and LibreOffice and I use Calc almost every day without problems.

Also, on the infrequent occasions that I need to do document recovery (usually from my unplugged laptop running out of battery), it works flawlessly.

Not to pick on you, but I feel like comments like yours are, I don't know, unnecessarily negative or nitpicky? Like, people will read your comment and think, "Oh geez, LibreOffice is crap." But it's used by portions of government on at least 754,000 machines [0] and my mom. Being used by many doesn't mean it's amazing, I suppose, but it at least means it generally works and probably isn't complete crap.

[0] https://www.libreoffice.org/discover/who-uses-libreoffice/

Do people use LibreOffice because it's open-source or because it you don't have to pay for it? In either case, the "enterprise edition" they envision will fail to service either camp.

If you have to spend any money at all, you might as well spend the money that MS Office costs. Or else, the main competitor is not MS Office but Google Docs.

The idea is to entice Enterprise and similar customers to pay for support, spending a small fraction of the money they save from not buying MS, to help keep the LibreOffice eco-system going.

So still much cheaper than MS :-)

The people I know use LO for essentially political reasons.

I've never really tried to use anything other than MS Office, because for my whole career I've been interacting with people on Windows using standard Windows/MSFT tools, and so any inconsistency / glitch in document exchange would automatically be my fault if I insisted on using LO or Apple's tools or whatever.

Plus, honest go God, Word and Excel are really, really good at what they do. Word got their more slowly, but once Excel ate Lotus 1-2-3, there really never was another competitor there, and Excel just kept getting more and more powerful. It's a wonderful tool.

(Now, there ARE people who learn Excel but refuse to go further -- into a true database, or into a proper business intelligent /reporting tool, or whatever -- and end up creating their own really janky versions of these things within Excel with macros and insane formulas hidden out in AA:5234 or whtaever; that's a problem for sure. But it's not so much an Excel problem as it is a problem with the lack of an obvious next-step ramp for those users.)

>But it's not so much an Excel problem as it is a problem with the lack of an obvious next-step ramp for those users

This is something of a general issue that was essentially created by the success of MS Office. With "everyone" using MS Office or something quite similar, there really wasn't a huge market or interest in a different program that was just an incremental step up.

There are exceptions but, for the most part, people tried to push MS Office to do things it wasn't really designed to do because the next step was jumping to some complex and expensive program from Adobe or whoever.

> Now, there ARE people who learn Excel but refuse to go further...

Since Power Query was first included by default in 2016, I've introduced tens of regular Excel users. Because there's no way to work cell by agonizing cell, there's a mental tension for a few minutes then usually the concept of working with fields and sets clicks into place so well you can practically hear it. Even if they never use PQ again, they often change how they use spreadsheets for the better. If you're really lucky they don't just start using PQ, they find the advanced editor and start doing some basic fiddling around. While it's fair to criticize M [0] as a niche language that only exists in a tiny corner of one company's data modeling features, it's also what finally motivated me to learn functional programming. That accessible step by step ramp from introduction to functional language data modeling in a common office tool is an underrated juggernaut of a feature.

[0] https://docs.microsoft.com/en-us/powerquery-m/

>for my whole career I've been interacting with people on Windows using standard Windows/MSFT tools, and so any inconsistency / glitch in document exchange would automatically be my fault if I insisted on using LO or Apple's tools or whatever. //

How long is that career? Over my working life StarOffice/OpenOffice/OpenOffice.org/LibreOffice have been far more compatible (for relatively basic documents) with other versions of MSWord than MSWord has, I don't really have experience of cross-compatibility from the last 5 years or so though.

30 years.

And no, what you say absolutely has not been my experience. Any rich formatting tends to get lost or munged, and support for stylesheets / templates between Office and non-Office platforms was woefully lacking.

If you're alone, sure, the free or Free option might be a good choice. But your time has value, so if you're collaborating with other people using true Office, just use Office. You'll be happier.

I use LibreOffice because it runs on my machine - MS Office doesn't.

If MS made a Linux version I'd probably get that.

LibreOffice is at its best if you make the docs and edit them just in LibreOffice. It's great at making structured docs. Calc is pretty good too. The PowerPoint equivalent is the weakest of the bunch, crashing frequently.

However in the biz world you end up sharing docs and sheets with MS office users and that is where things start to go sideways. Complicated MS docs are not rendered well. Excel sheets are generally Ok. PowerPoint docs don't round trip at all well so I've given up even trying!

I really wouldn't be surprised if this happens soon. Some kind of encrypted flatpak with some licensing scheme built in.

Or making a Linux store of essentially an apt server where everything has a pricetag. It shows up in your repository after you purchase it and then can update it like anything else... This isn't new ground these days

There's lots of ways to do commercial software with the existing infrastructure and I don't think the modern Microsoft cares as much as long as they are selling.

> If MS made a Linux version I'd probably get that.

What about the web based Office 365? It should run fine on Linux, and clearly paid closed-source software isn't an issue for you if you would be willing to pay for traditional Microsoft Office.

> However in the biz world you end up sharing docs and sheets with MS office users and that is where things start to go sideways.

Office 365 should work swimmingly for this use case.

Had to use Office356 recently, because libreoffice kept putting strange, large, unremovable rectangles in the spreadsheet I got handed. I wished, i would not have to touch this Microsoft filth, but I had work to do and libreoffice kept bugging.

Personally £7.99 a month for a piece of software I use 1 hour every 6 months is definitely not worth it. Though I'd pay £100 for it no questions asked.

Subscription plans are stupid.

> Though I'd pay £100 for it no questions asked.

Office 365 is a continually updated and enhanced product. They won't provide those updates and enhancements to you forever because your £100 won't pay for them.

What you are asking for is their old downloaded/shrink-wrapped Classic Office product offering, which you can still get by paying a f themixed sum ($145 I think).

But even in the old days, your £100 bought you a single program with very few updates if any. If you wanted new features or even security updates (when this became a thing), you needed to purchase a new license, for another £100.

What if you don't want the new features or security updates? Well, perhaps it sucks, but the world of software is a far more complicated than it used to be and both the feature competition and security threats are steep, so no software vendor can afford to sit on and continually support an old version. Unchanging software isn't a product that becomes an antique like a 1959 Chevy. Rather it loses value over time as the world changes around it and it becomes less relevant.

Eventually, they will stop updating your purchased software, and if you want to have new features and security protections, you will need to pay again. In the long run, that cost will converge with that of a subscription.

This is why some people like open source software like Libreoffice as an alternative model, but as the OP highlights, it still needs an economic model to incentivize the enhancement of the software, and it would appear that the primary motivation of many users of the product is to not pay for an office software suite, not the higher ideal of a shared community software development model.

I get what you mean, but people still made a living selling software without monthly subscriptions before. If I had to use Office more, I'd pay the monthly price, but not for 2 hours a year. It being continuously developed still doesn't justify the price.

I can afford it, but if I had to pay £7.99 x 6 = £47.94 for every hour of productivity I extract from a piece of software, I'd go bankrupt very quickly. That's why I called subscription plans stupid: it's unreasonably expensive for the amateur (like me) slice of the market.

And being such a small time user, I don't care about the super new features of Office $CURRENT_YEAR. If I could buy Office for Linux 2020 for 100 pounds, I would, and it would be enough for me for the next decade, unless I'll move out of software engineering and go work as a lawyer or something.

> I can afford it, but if I had to pay £7.99 x 6 = £47.94 for every hour of productivity I extract from a piece of software

Why would you pay for 6 months of subscription for every hour you use it? If it's just an hour of use, you just pay for 1 month, so 7.99. Is there a 6 month minimum?

> If I could buy Office for Linux 2020 for 100 pounds, I would, and it would be enough for me for the next decade,

I get that that would be a great deal for you, but why would MS want to support your £100 Office for Linux purchase with security and feature updates for 10 years? Why even develop and release that product? It's not like they lack for people who will pay their Office 365 subscription fee. It would make as little sense for them to do that as it would for you to sign up for that subscription.

At our research institute most people use Linux + LaTeX, which is great, but interfacing with MS Word is an absolute pain when you work with external collaborators. Accordingly, one of our professors‘ key argument for Mac OS is: „It’s a Unix with Word...“

Is there a manifesto or document anywhere which lays out the economic case for (US) Federal grants to develop free software? How many developers and how many projects could be staffed full-time with a billion dollars set aside annually for an Office of Open Source Software?

(Of course such an office would descend into a cluster-[bleep] of political warfare over topics that have nothing to do with developing software so spending nothing probably gets as much done in this area than creating a government office which costs money and doesn’t get anything accomplished.)

A better deal would be Fed paying MS, then MS charging other countries all over the world and MS collects big $$$ in profits and US tax paid. Much of SV started like this too.

> Another pathology is that there are companies who ship LibreOffice, often claiming support, but then file all their tickets up-stream and hope they are fixed for free.

Lmao... but thinking again, given the average user's bug-filing abilities, a professional bug-filer might actually be useful.

Indeed. I created this initiative some time ago, but sadly have not been able to get folks responsible for deployments involved: https://wiki.documentfoundation.org/QA/Guidelines_for_public...

I guess it's just another reflection of the passivity we witness everywhere.

I am looking forward to future development of LibreOffice. For me it has become better and better. I've never had a new installation, or update of LO which made me think "If only I had stuck to the old version!"

More and more I am using plain text formats these days, which I then convert into other things, but from time to time it is nice to have LO at the ready, knowing, that it is free software and as such will always be available, in our collective resources.

A friend of mine recently complained about pictures disappearing and floating over text area margins in their document and having to put them back in in a 200 page document. I've never had such a problem. However, there seem to be some rough edges left to improve.

I cannot complain about the user interface actually. To me it is quite intuitive and I find everything I usually need rather quickly.

It seems like this whole thing could be solved by renaming the tag from "Personal Edition" to "Lite Edition" or something. People want to use and "office" product for work, duh. The document foundation keep arguing that funding is needed and the users keep arguing that they want a free version for commercial use and that naming is important because "the boss won't read the license agreement but look at the name! Personal Edition"

Why are people on the defensive always so blinded by the obvious? If you understand what people are getting upset about then, on the first line of the letter, they should state, unambiguously, exactly what people are having an issue with. Instead, the introduction is about the success of LibraOffice which is not contended and actually somewhat irrelevant. Then this:

> So it is a bit surprising to see the project's core developers in a sort of crisis mode while users worry about a tag that showed up in the project's repository

This sentence just tells the reader that you still don't understand what the fuss is about because you are referring to past good will and not the simple fact that the name is wrong. Readers want to know that you understand their point of view before they are willing to read about your point of view.

"Lite Edition" isn't much better. The usual name for this kind of thing is "Community Edition", which has no negative connotations. But I don't see why it needs any special name. The LibreOffice brand is already harmed by the continued existence of OpenOffice, which many people still think of as the default free (as in beer) office suite despite its obsolescence. Complicating the LibreOffice name will only make things worse. IMO it should remain plain "LibreOffice", and the paid support version can be "LibreOffice Enterprise Edition".

Someone from TDF responded to a similar comment on Reddit:

https://www.reddit.com/r/linux/comments/ho0n51/update_on_lib...

I just loaded up LibreOffice to give it a try, I would like to move off of Google as much as possible. 5 minutes in, I'm like, this is not that bad! Then this happened:

> Due to an unexpected error, LibreOffice crashed. All the files you were working on will now be saved. The next time LibreOffice is launched, your files will be recovered automatically.

I really require tools with multi platform live collaboration--I've been looking at Apple's Calc, as a possible switch, as well.

Maybe I'm seeing what I want but it seems that the plan is for the document formats and editing programs to be free while generating revenue from infrastructure around delivering something "Sharepoint-like" (or a SaaS service to do similar), complete with a throat to choke/vendor to blame if things go sideways. I don't see this as fundamentally different from the choice of running CentOS versus paying for RHEL: the latter is the same software but comes with enterprise assurances that allow the CIO to sleep at night.

I'm sure there's someone could write a compelling essay along the lines of "Free Software isn't free of charge" if someone were inclined to pursue it. Even people like RMS have to eat from time to time.

> The problem is compounded by companies that sell inexpensive "support" for LibreOffice, but which are not involved in its development and are not really able to provide that support. Those companies "file all their tickets up-stream and hope they are fixed for free". Companies working in that mode have no problem pricing their offerings below those of the companies doing the actual work (and thus winning much of the business that does exist). In addition, they simply call their offerings "LibreOffice", which actually looks more authentic than services from other companies, which are trying to build their own brands around LibreOffice support.

Is this not a cut-and-dry case of trademark violation?

Every time I open LibreOffice and see the loading logo... I'm amazed it's free software.

I'm thankful to everyone that has contributed to it.

The Document Foundation should - IMHO - employ a more nuanced, and "deeper", community funding effort. Even though, when downloading it, we see a payment link - there is almost no motivation to press it:

1. We are in a hurry to go on with the download.

2. It's called a "donation" - which it isn't. It's support for the project, financing further development and bug removal, and should be presented that way.

3. We have zero information about the situation The Document Foundation is in, what it needs money for, what its income stream look like and why, etc. I think Thunderbird (https://www.thunderbird.net/) has been doing a better job in that respect.

4. There doesn't seem to be an effort to encourage or even make-it-easy-for users (corporate or individual) to engage in community activity, of the kind that increases commitment to the project and also awareness to the importance of supporting the project financially.

------------------------

Also - please consider actually supporting both projects I mentioned, financially:

* LibreOffice: https://www.libreoffice.org/donate/

* Thunderbird: https://give.thunderbird.net/en-US/

and consider doing some volunteer QA work on LibreOffice by reporting bugs or even getting into the code. I, particularly, am very involved with Right-to-Left, Hebrew and some Arabic language issues.

Instead of delivering the dead and long stalled project Apache should admit the failure of OpenOffice (with their commit rate vs LibreOffice's one it's obvious) and redirect to LibreOffice instead. They should officially deprecate it. Right now they are stealing potential users of LO with their substandard product.

I haven't used Libreoffice in years.

Did they ever update the UI styling?

I remember it had a very "java" look to it.

You know that feeling when you open an app, you know it's going to be a bit janky.

Do you happen to be on Mac? On Debian-based systems, AFAIK, there are specific packages for DE integration, for example, for GTK3, which makes LO look native (although, I think that the styling is highly customized, because of the functionality requirements).

You can configure it to have a ribbon, and other complex toolbar layouts too. I use it for work on Linux, and I think it's quite well done - and I hate Java also.

It has visual class libraries that give LO a slightly different look depending on what you're running it on.

I'm typing this from a Debian system running KDE Plasma, and my LibreOffice reports using the gtk3_kde5 VCL. It uses Qt5 widgets and looks, for the most part, like a regular KDE Qt5 application. It no longer looks or feels like a Java application. Because it's not - Java is optional except in LibreOffice Base. You can disable JRE in the options, and LO parts that are not Base will retain nearly all functionality.

To be completely fair, it never looked like a Java application.

>You know that feeling when you open an app, you know it's going to be a bit janky.

I've used Electron apps before, yes.

GTK3 has the best integration, it now uses native widgets and elements thanks to the work of Caolán McNamara (it is hard to comprehend the amount of effort required). Example: https://www.ubuntubuzz.com/2020/06/libreoffice-looks-gorgeou... (this is using the tabbed interface from View - User interface)

> so nobody thinks in terms of buying support for any office suite[.]

In my experience working in a repair shop, people only think in these terms when buying an office suite, regardless of if they'll actually use the official support. This only applies to individuals though, not corps.

I‘d like to use a open source text editor. But libreoffice text rendering is so bad, i cant use it. At least thats the case for me on osx, but i think this applies to all plattforms.

Just compare text rendered in libreoffice and to the one in pages. It hurts my eyes.

The upcoming version is moving from Cairo to Skia for text rendering. It might help.

(LO Dev) Unfortunately, mac has the least interested programmers available to debug issues, and there is some weird interaction going on with our rendering on macOS that we can't track down.

They do a great job.

But I use [commercial] Softmaker Office. They offer a free version too: https://www.freeoffice.com/en/

Thanks for posting this.

I'm looking for a MS Office replacement, not heard of Softmaker Office before. I'm still using an old version of Office, but it doesn't work well on a high DPI screen.

What's your experience with Softmaker Office - anything you're missing from MS Office (if you've used it before Softmaker Office)?

I miss nothing. Buts thats me. If you need visual basic scripts, maybe it is not for you. AFAIK it has its own scripting language. There is no risk to test the free version. You can actually use the free version, the limitations are not tremendous. I always buy it.

Personally I find LibreOffice Writer to be fine for my regular reports until something goes wrong and then it's like old versions of Word.

For example, occasionally when I'm making a numbered list the tenth item and beyond will be bounced over a full tabstop which looks horrible. Then I'm trying to work around the problem by messing with the tab stops which causes problems for the rest of the document.

It's really hard to fix when something goes wrong as a casual user. Inserting images/figures is fraught with peril too. I sometimes find myself hacking the doc until it looks presentable and then not touching it. It's like those old "smart" systems that weren't quite smart enough to get it right, but would fight you tooth and nail when you tried to undo what they did.

For what its worth I prefer the consistency of its interface over the ever changing ribbon of Word. Word also tries to guess what you want and when it guesses wrong it's really annoying.

We kind of need a version of CC-BY-NC. Preferably with a governance model that then distributes money from commercial licenses to contributors... Or government funded Open source. There is no reason for society to try to do something as intrinsically cooperative as building software through market mechanisms.

Summary"

... There is the Personal Edition, which would be "forever free" and only available from the Document Foundation. This release would be tagged, according to the plan, "volunteer supported, not suggested for production environments or strategic documents". The alternative would be "LibreOffice Enterprise", which would only be available from "ecosystem members". This version would come with commercial support and a corresponding price tag. ... "There will be an X month gap between the release of the two versions: LibreOffice Online Enterprise and LibreOffice Online Personal".

Have any of the support companies tried paying bounties for tickets? Seems like it would get revenue pointed at the bugs/enhancements that customers wanted worked on.

Do they have a subscription to pay LO directly?

I don't mind LibreOffice's interface in general. Only a sublime style Ctrl+p find would make a huge difference. That alone would put it's interface on top of MS office in my opinion.

I find it noteworthy that LibreOffice Online is what's bringing in cash according to TFA. What it might tell us is that the prevalence of cloud and online services today is just a consequence and natural market reaction to F/OSS proliferation. LO/OO.o themselves have open-sourced their product many years ago when it still was owned by Sun (I even know developers in Hamburg who worked on it in the 1990's when it still was called Star Office).

I have a few questions that don't seem to be fully answered in the post.

> This “Personal Edition” tag ... has the purpose of differentiating the current, free and community-supported LibreOffice from a LibreOffice Enterprise set of products and services provided by the members of our ecosystem.

Will this "LibreOffice Enterprise" be Free as in Freedom? Would I be able to package it up myself and call it "Community ENTerprise Office Suite"?

These enterprise versions are already available since years, for example:

https://www.collaboraoffice.com/solutions/collabora-office/

https://libreoffice.cib.de/

This isn't really changing anything. LibreOffice is under the GPL. It's just a marketing attempt to get large organizations to realize that they should contribute back to the ecosystem if they rely on LibreOffice.

LibreOffice is currently under the LGPLv3 and the MPLv2, not the GPL.

Further, the document refers to 'new products' that could be under the 'LibreOffice Enterprise' umbrella. Could any of these be proprietary? Do we have any assurances?

You are correct. I don't know how I fabricated that idea.

In response to your second point, there seems to be strong opposition to open core from some key players in the community.

Personal Edition implies it won't have important collaboration features needed in organizations. And whether that will be true or not isn't the issue, less volunteers and non-profits will end up using LibreOffice, instead they could just called it Community Edition like every other FOSS project. They will end up with less volunteer contributors, but maybe that's what TDF wants, it's a pity.

Can we expect an Auto-Update within the next 5 years?

What's wrong with updating via Flatpack or AppImage (or Brew or Chocolatey)?

About people hating change, in my IDE an update renamed a menu from "revert" to "Rollback" , for a while I had issues finding this menu because my brain was searching for the old text and just skipping over the new one.

Is it my fault that change causes productivity loss? Should we take a pause and train ourselves after each software/webpage does a change?

Kind of odd that they don't compare themselves to Blender and its successful foundation in the article, not even mentioned once.

Is there a hosted service for LibreOffice online ? I wish I could find a better replacement for Google Drive.

Actually read the article...

They say they don't host the service.

I recommend checking out ONLYOFFICE. They have a personal tier with a small online storage for free. And everything is AGPLv3.

https://personal.onlyoffice.com/

OnlyOffice have shot themselves in the foot when made it use MSO file format to store data. Should have build on ODT

OnlyOffice are not a community-oriented project, with their aggressive usage of AGPL+CLA, previous policy of withholding parts of code and throwing the rest over the wall once in a while, and their compatibility is far from what they claim: LibreOffice is far better compatible with MSO. Also, feature-wise it's crippled in comparison not only with LibreOffice proper, but even with LibreOffice Online.

> LibreOffice is far better compatible with MSO

Why do you say that? I trialed both of them with an array of complex Word documents and OnlyOffice far outperformed.

> > LibreOffice is far better compatible with MSO

> Why do you say that? I trialed both of them with an array of complex Word documents and OnlyOffice far outperformed.

Not in my experience. OnlyOffice often failed to render simple documents LibreOffice had no problems with at all.

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

I was hoping to find an actual plan towards making the software better, more usable, and desirable. Instead it’s just a “marketing plan” about splitting apart something that lags behind.

Where’s the vision?!

So much talk about UI this and that - how about leaving UI and feature set alone and instead focus on speed and memory foot print?

Though recent versions are better, it still feels slow compared to Office.

If LibreOffice is offering a paid version for extra features / services, considering what they are already offering, I would certainly paid, or at least have a strong interest.

Wordstar and Wordperfect were easier to use imo.

They also missed a lot of features. And IMO not needing the 'reveal codes' screen and having to clean up markup codes manually is the reason Word won the wordprocessor 'war'.

I still remember the inconveniently tall keyboard templates (e.g. https://www.google.com/search?tbm=isch&hl=en&source=hp&q=wor...) for WordPerfect to tell users where all the function were. Give me the ribbon bar anyday.

GUI was a mistake.

so basically iceweasel vs. firefox again. I'm fine with that (and still miss the icon)

Could be that the enterprise will include some cloud sync for auto backup (like O365/OneDrive), of course that will not be free. Best case (call me naive) but it could just be that the personal edition simply does not ask for the syncing, worst case, it will nag you constantly.

well, if you're paying for the cloud, you can pay for the free-as in freedom-software as well I guess ;)

Is there any electron gpl based Word and Excel alternative? I always imagined running on top of electron helped you skip a lot of work and helped the app not look like shit/LibreOffice

You know what helps skip a lot of work? Not needlessly rewriting things.

Libreoffice looks like a native desktop app. How do electron apps look better than that?

Plus, it is possible to customise the icons and colour scheme. There's light mode, dark mode, blue mode, even pink...

Depends on which desktop. On macOS it stands out like a sore thumb. Icons are almost monochrome and indistinguishable. Also, it does not feel as smooth as a native Mac app since it uses its own toolkit instead of Mac native scrolling etc.

If you want to see how moving from Java to Electron destroys an application, while making it look "better", try out GeoGebra Classic 6 and GeoGebra Classic 5.

The difference is stunning. Geogebra 5, written in Java, is just so much more performant, and feels so much more like an application, not a broken website.

The same actually applies to pgAdmin 3 vs pgAdmin 4.

pgAdmin 4 is not an electron app. It is a real web app, written in Python and Flask, and Qt browser engine bundled together.

Qt's browser engine is the Chromium Embedded Framework, Electron is the Chromium Embedded Framework.

Sure, the backend is different, but in a thread about UI issues, it's the same.

The UI issues in pgadmin4 are caused exactly because it opens web pages from relatively slow local web server, synchronously.

Electron/CEF can be used to make a decently (i.e. usable) performing SQL client app (see Azure Data Studio). But not in a way that pgadmin4 did it.

There is, actually. ONLYOFFICE offers their AGPL word processor, spreadsheet etc programs as Electron apps too. The UI is absolutely modern, and one of their other selling point is their DOCX/XLSX compatibility.

https://www.onlyoffice.com/desktop.aspx

> selling point is their DOCX/XLSX compatibility.

Which they fail to deliver, given that LibreOffice manages to be more compatible with docx than they are.

Reply on news.ycombinator.com