UI is For Users & Waffle Menus are for Diners

đŸ’» Development
, , , ,
1

Dear design developers,

I’ve run the self-hosted Nextcloud server & desktop client since before it was even named NextCloud. In production. Every day. So my hope is this lands as feedback from someone invested in the project rather than a drive-by complaint.

The recent addition of the app-menu grid (“waffle menu”) to the web & desktop client is a step backwards. As such, I would like to explain why in a way that’s useful rather than just grizzled old grumpy curmudgeon, which can unfortunately be my default.

It turned one click into two. Opening a folder used to be a single action. Now it’s a menu, then the item. That’s a small tax, but multiply it across every user, every day, for years, and the end users spend a lot of collective time to gain nothing ever asked for out of the software.

Consider what you’re competing against. The MacOS Finder bar is visually & functionally the same as it was 40+ years ago. So whether you learned in 1984 or 2024 it is the SAME, that is not stagnation — it is the highest compliment a piece of software design can earn.

MacOS-FInder-Bar-1984-2026
MacOS-FInder-Bar-1984-2026900×500 15.6 KB

Four decades of muscle memory that never had to be relearned. Millions of users who never filed a support ticket asking “where did my thing go?” That is the bar. Interfaces that do not change are interfaces that do not generate confusion, do not break workflows, and do not cost administrators a support queue every release.

Burgeoning Problematic Attack Surface

Instilling a fear of ‘well this update is likely to foobar my workflow’ is how you wind up with an ever growing list of outdated builds in the wild, just waiting to be compromised and 2FA ain’t going to save you.

Why update if it means an influx of support tickets, CSS hacks, and other workarounds just to get back to where you were prior to the upgrade? Check the books on the insane number of People remaining on Windows 10 with some still clinging to air-gapped Windows 7 installs to run machine shops.

The people who feel this most are the ones running it in production — the admins supporting non-technical users, the shops where every UI change means re-training and a fresh round of “hey um it looks different, is it broken?” tickets. Novelty in a sync client isn’t a feature to us. Predictability is the gold standard.

I sincerely love this project and have bet on it over the Spyware from Surveillance Valley in real infrastructure for over a decade.

All I am asking (as well as many others according to search returns) is that “we redesigned it” be held to a higher standard than “we could so we did.” Changes to core interaction earn their way in by solving a real user problem, not a design-team itch. When in doubt, leave the click count where it was. Or at a minimum introduce it as an option.

Sincerely, a long-time user who’d like to still be here at version 50,

Allison

###

2

Hello @LibertyTalk,

you started a topic in development category. This category is intended for active developers of the core or apps in the Nextcloud ecosystem.
From the description in your topic, it seems as if you are not directly seeking help and advice about a concrete problem you have, but want to give feedback to the developers and designers.

I will forward this post to the design team. Not sure in what form they might react.

Especially, if you want to report a concrete issue for the corresponding developers to address, this is unfortunately not the right location: most devs will not likely stumble upon the issues in the forum. Such issues are better tracked on GitHub in the appropriate issue tracker. Probably, this is the issue tracker of the server but could be another one as well.

Regards,
christianlupus

3

There are already several forum threads and GitHub issues discussing the waffle menu. Instead of writing a long essay that basically just says, “I don’t like change. :wink:”, it would probably be more productive to join those discussions with constructive suggestions on how the waffle menu could be improved or extended.

Just my two cents: I don’t think the old menu was the ultimate solution either. It had some objective drawbacks, especially on instances with a large number of installed apps. My preferred solution would be to enhance the waffle menu by allowing users to pin their most frequently used apps to the top bar as favorites. That would give us the best of both worlds.

Apart from that, the waffle menu, and its corresponding icon, is already a well-known UI pattern. It’s used, for example, across Google services, in a similar form by Android’s app drawer, and, more generally, by virtually every operating system that provides some kind of application launcher or Start menu. So I think the argument that a UI change like this requires ‘retraining’ is a bit of a stretch. :wink:

4

I am annoyed by that, too. :wink:

So should you because Finder definitely is not what we are competing against. :melting_face: That comparison does not work like the famous idiom of comparing apples to oranges. Finder was not perfect from its inception, required iterations as well but most importantly is a file manager with a well defined scope of features and purpose that is essential on most platforms, defined already long ago. Nextcloud server on the other hand is a collaboration platform by now (no longer just a file exchange)



and this is one of the key reasons why the old menu did not work anymore. Not even for ourselves at Nextcloud GmbH. So, better compare the waffle menu to the Window start menu with its many iterations or Launchpad[1] on macOS or Spotlight or whatever you use to launch apps. Also, Nextcloud itself just turned 10[2] years old. The whole field we are active in started to exist only in the past 20 years. File managers like Finder were basic computing needs figured out already before I was born. :baby:

Another important argument made, especially in regard to “Consider what you’re competing against” - exactly! The similarity to the leaders[3] in our field of competition is a key factor of user acceptance and adoption. Without adoption, there is no growth in user base. Without growth, change in the centralized, siloed landscape — as one of the key motivations — is unreachable. And overall, we are developing software instead of just preserving it. Naturally, not every change turns out be an actual improvement. Though there always is a chance to make up for it.

We are because we cannot pay bills by scratching design team itches. :sweat_smile: I mean, that is the whole business model. :blush: In the end I just wanted to shed a different light on this. I was not involved in the design process of these changes but still I have insights which probably are not there looking from the outside. Certainly, things are not changed just for fun or make them look fancy.

:bouquet:

  1. Launchpad was introduced with Mac OS X Lion (Version 10.7) on July 2011. In context of the earlier comparison of Nextcloud server to Finder, I just wanted to point out that this type of user interface has been around some time already as well. The naming as “waffle menu” is the only new thing about it. ↩

  2. A few years more, if you count in the predecessor ownCloud from which it was forked. ↩

  3. We are not there yet but working on it :wink: ↩

5

Me too, I’ve been a user since ownCloud 4.

Agreed. If I had 10-20 apps installed, I’d probably appreciate it, but visually hiding my five apps behind an additional click is annoying.

This should be an option that we can turn on or off. (Or at least default to waffle only if you have seven or more apps.)

Yep, that’s called stagnation.

It really is stagnation, and the highest form of compliment is imitation. No one is imitating your 42 year old design, not any more. It was a practical design to copy 30 years ago, but was already dated and being depreciated by the Vista days.

“40 years of muscle memory” is a very poor excuse. I’ve been using computers since 1982, I even had a Lisa, but I was somehow able to move on when better UI options were available.

I particularly love the search introduced with Vista, which was improved with 7 (and then diminished ever since). I no longer had to rely on muscle memory to go through level after level of menu hierarchies.

Ah, that’s why it was unfamiliar. I’ve lived google free since about 2010, I literally use Nextcloud and other OSS so I don’t have to use google malware. Copying google UI/UX is not the flex you’re implying it is!

I really hate having my options visually hidden.

I broadly agree, I really do, but many users can be really thick. I’ve had colleagues look at me like I was a wizard because alt-tabbed between applications rather than clicking their icon in the taskbar. I had another that was convinced I’d installed third-party software because I selected a paragraph by triple-clicking it.

Every UI change causes disruption, some more than others. I don’t think this change is a big one or that it will impact many users, but I really hate the “visually hide everything” default.

6

Didn’t ownCloud introduce a waffle menu back in ownCloud 8? As far as I remember, they have kept it ever since. :wink:

Before that, the apps were listed in the sidebar, which wasn’t exactly the ultimate UI solution either.

You know, two things can be true at the same time. Software and services can be privacy-invasive while still offering excellent functionality and a good UI/UX. In my opinion, that’s largely true for Google Workspace. Whether the waffle menu, as Google implements it, is one of the good parts, I’ll leave open for debate. :wink:

Either way, I’ve been using Nextcloud since version 10, and I’ve been self-hosting FOSS software and NAS appliances for even longer. I’ve also been using Linux as my primary desktop OS for about 15 years. However, a supposedly superior UI/UX was definitely never the deciding factor for me to switch to FOSS alternatives, especially not 10 or even 15 years ago. :wink:

That’s exactly why I think the ideal solution would be to let users pin their five or so most frequently used apps to the top bar next to the waffle menu. Best of both worlds. :slightly_smiling_face:

7

I don’t remember, I haven’t used ownCloud in over a decade. I progressed, I moved on a long time ago, I’m not the one wishing I could use 42 year old muscle memory :stuck_out_tongue:

I didn’t say otherwise, I just said I lived google free since 2010 so didn’t recognise the google UI. You seem to be reading a lot into what I actually said. I’m not the OP, I definitely don’t hold their “get off my lawn” views.

I simply don’t like hidden by default. I even set the Windows taskbar to “Never combine”.

Me too, I’ve been self hosting email since 2004 and I’ve run a samba NAS since 1997.

Now that’s masochism! I used linux from about 1994 (first slackware, then debian from 1997), but was well and truly over it by 2004 when I switched to FreeBSD. So much better, more consistent. Self hosting was a breeze on FreeBSD compared to linux.

I’m jiggy with that! We’ve found common ground :smiley:

8

Yeah, maybe I misunderstood your point. My reply was specifically about “Copying Google UI/UX is not the flex you’re implying it is!”.

Yes, Google is evil. And yeah, copying someone else’s UI/UX might not be particularly innovative, but it doesn’t have to be if it works well. :wink:

That said, I still think a 1:1 copy of Google’s waffle menu is a bit too inflexible for Nextcloud. It makes a lot more sense for Google’s ecosystem because, let’s be honest, they have dozens of apps spread across multiple categories. There’s simply no practical way to make all of them directly accessible.

On top of that, the Google workflow is usually very tab-oriented. You have one browser-tab for Gmail, another for Drive, and every document opens in a new tab. Also calendar, tasks, email etc works together more seamlessly. In day-to-day use, you rarely need to launch any apps directly.

Yeah, I don’t think we’re actually that far apart. And lot of those things also come down to personal preference and habit. I’m generally more of a minimalist, distraction-free kind of guy. But in this particular case, I’d definitely like to have my most frequently used apps just one click away. :grinning_face_with_smiling_eyes:

Btw, as a workaround for now, I guess we could use our browser’s bookmarks bar and pin our most frequently used apps there. :wink:

9

A few more general thoughts:

One problem these days is that things are often replaced rather than extended. With Windows, for example, many UI/UX features have been added over the years, but up until Windows 11 (and the disastrous Windows 8), you could usually ignore most of them because the old workflow still remained available.

Nowadays, changes are often introduced that solve a real problem for a specific group of users, but then the new workflow is forced onto everyone, including users who don’t actually need it.

Ideally, the Waffle menu would either have been introduced as an additional option while keeping the existing approach, or the new menu should only have been released once it included something like a favorites bar. That way, users who prefer the old workflow, which also has some objective advantages, would not lose a way of working that already suited them.

That said, this is often a question of resources, which, especially in FOSS projects, are limited. So, I suppose that, at some point, compromises have to be made and decisions have to be taken in the knowledge that they won’t please everyone.

10

Well it’s early, Canada is poisoning the skies with their incompetent forest mismanagement, so feeling a little surly and just utterly mystified by some of these responses.

So why like the Finder, can it not be a combination of the Waffle and say like the four most common access-points that were there from the start?

Files | Photos | Activity | Notes | and then hide all the other collaborative tools behind the waffle? How is this difficult?

Or MAYBE, let users select their 4 most common touchstones and hide the rest of the collaborative tools behind the waffle?

Old adage: Dance with the girl you brought, i.e. stick to what got you to the table. Seamless file sharing.

Reminds me of a Steve Jobs video on latest tech, developers & a Steve Ballmer meme.

Moreover “We have to eat” sentiment of development thus saying “Pound sand” to the actual users who now deal with borked up workflows, yeah great way to sell a product.

Also, interesting no one addressed the security aspects of pushing users into not updating as a means to avoid getting their workflows burned to the ground.

11

That could certainly work. But what if I’d rather have Calendar or Memories there instead of Activities or Photos? That’s why I think a customizable favorites bar would be a better solution than hardcoding a handful of apps.

That’s basically what I’m suggesting as well. Although I’d be perfectly happy if it supported more than just four favorites. :wink:

Well, I think we’ve long since moved beyond that, and I actually think that’s a good thing. I don’t use dozens of apps either, but I certainly wouldn’t want to give up things like Calendar, Contacts, Tasks, Office, and a few others.

If all you need is simple file sharing, there are plenty of alternatives available. And in that case
 what’s really the problem with the waffle menu? I mean, realistically, if file sharing is all you use, you’ll probably be spending 90% of your time in the Files app anyway. :wink:

At the end of the day, that’s a question for the developers and designers. Personally, I think it’s a bit unfair, especially with FOSS projects that we get to use for free, to assume that every requested feature should simply exist or can be added “just like that.” Resources are limited, so priorities have to be set, and every new feature comes with development, maintenance, and long-term support costs.

I’m sure they considered the trade-offs. One of the stated reasons for the change was to make Nextcloud feel more familiar to users coming from Google Workspace and similar platforms.

Again, I think that’s a bit of a stretch in this particular case. If someone chooses continue to use EOL software because of a UI change like this, that’s ultimately their own decision. And for larger organizations, Enterprise subscriptions with LTS releases are available.

Overall, I do share the general sentiment that there’s room for improvement here. But I don’t think this is the kind of change that suddenly makes Nextcloud dramatically harder to use, requires extensive user training, or causes a major productivity drop. Is it a little less convenient for some users? Sure.

Either way, I think it’s more productive to keep the discussion grounded and focus on concrete improvements, for example by opening GitHub issues, upvoting existing ones, or contributing constructive feedback, :slight_smile:

12

I miss the old internet and common sense.

“Yea just go use something else.” Another classic circa 2026 sales pitch.

The Ivy League refers to such as ‘Entropy’ although my preference is for the more unvarnished mainstream slang.

Good grief Charlie Brown.

13

I already pointed out earlier, but the concept of the “waffle menu” is not new and not specific to Google around which the whole discussion circles all the time. It is a really common pattern in user interfaces. I want to point out again: before Launchpad, there was the grid view for the “Applications” folder in the Dock on macOS in 2010 when I bought my first Mac. The only new thing is the “waffle” name of this concept. It is a proven solution to an age old problem. :woman_shrugging:

The community forum is the wrong place for this sort of discussion because most people around here are not actively involved in the development. If you want to get involved in our design and user experience process, please take a look at our website. The contribution and design pages in particular. We invite everyone to contribute. It is an open project with a lot of community input after all. I even presented a few major cooperative efforts in our latest major release of Nextcloud Talk in our Nextcloud Hub 26: Spring release video.

I see that the misconception of selling software persists. We do not sell a product or software. We sell services around it. Mostly: customization and support. We cannot sell what people can simply take for free. That does not work. Open source licensing requires different business models. And we cannot challenge big tech and change the landscape without appealing to their users and the majority over all. With its dynamic, there will always be someone disappointed. Overall change is innate to all things computer. Even Finder. :wink: The introduction of tabbed windows in Finder was the greatest change, by the way. :relieved_face:

And it is not a “sentiment” but a matter of fact that – yes – we have to live off something as well. If we cannot maintain a project like this in a profitable way that allows people to be paid for working on it, then there will barely anyone left to complain to (probably about the even more severe security implications you mentioned because no one is paid to properly take care of them, see the whole Heartbleed story from 2014 which also rooted in a lack of funding). When developers cannot earn a living from working on or around a software, they need to find other jobs. That means: dramatically less effort invested in such a project. Hence the requirement of steering such a project responsibly and with reason and at least economic self preservation.

Nextcloud does not push anyone anywhere. :sweat_smile: We are not even in the position (unlike big tech on which whole nations depend on). You can stay on outdated releases, it is your choice (and responsibility). You can also take and adapt Nextcloud to your liking, it is free and open source software.

:woman_facepalming: Oké, I am out of here now. :wink:

14

Yeah, me too sometimes. Especially after reading posts like this one. :wink:

So, would you actually pay money for it if they brought the old menu back?

Otherwise, pulling out a single sentence while ignoring the broader context is also a pretty common 2026 internet phenomenon. It’s a tactic used by all kinds of trolls, conspiracy theorists, and political or religious fanatics to derail discussions instead of engaging with the actual argument.

15

Yeah
 I, on the other hand, am at liberty to say “shit” all day.

And I can tell you with confidence that this change is not what the term enshittification means in the context of software development or the tech industry.

16

Custom menu has been updated.

We can use that to return to the days before waffle menu

17

Yep. Would happily pay a reasonable sum for an LTS, documentation and a sales pipeline that was not utterly so obtuse/convoluted that winds up with some ‘US partner’ that has no clear way to simply purchase access. Or my real favorite, you ferret out & call the telephone number, it rings to a phone tree, press the number for sales it then rings, takes you to voicemail, and you get a return call 6 months after the project passed and you already found a work around.

Take my money PLEASE.

Here’s the order of operations: What does the customer/user want, then build the tech to meet that specific need.

Not: hey look what we can do so now let’s wedge it into the customer/user’s experience and bork their established workflow.

Or if you are Big Tech: Oh and let’s throw in some OTP/2FA/passkey/CAPTCHA/biometrics/Age Verification that turns a 10 second login to a 5 minute multi-app affair.

18

This discussion gets off-topic related to topic title.

19

Well, this is a community forum, so we can’t speak on behalf of Nextcloud GmbH or provide support for anything related to the corporate side of Nextcloud. I also don’t have any personal experience of working with them, nor do I know what their partner network in the US looks like.

Either way, for now, if you’re already on version 34, you’ll probably just have to live with the menu. If not, you can still delay the update for a few more months, even without Enterprise Support: Maintenance and Release Schedule · nextcloud/server Wiki · GitHub

Or you could try the Custom Menu app that @jaycedk mentioned: Custom menu - Apps - App Store - Nextcloud. I tested it briefly on a test instance, and it lets you pin apps to the top bar, so you can more or less recreate the previous navigation.

20

Thanks for all your feedback and help.