First if all, it wants to do everything and does none well (or better than specialized apps)

Its internals are a shitshow - a question about getting the real IP of the calling client raised a 10 pages discussion where people said "it works" and other days "it does not". The "solution" required you to change a volatile setting within the docker container (and do it again when the container changes)

Finally the upgrade is insane. I once tried to upgrade my installation, was told that n+2 upgrades are not possible (fair enough) and ... the installer destroyed the database. What a piece of crap. Thanks god for backups.

Not once had I any serious issues updating when pulling new updated images.

Occasionally it whines about missing indices, but that is easily fixed using the occ command line tool. The clients real IP is forwarded by the proxy.

What I want to say is just that Nextcloud works fine.

As soon as I started to add apps from the store it starts to be a pain. An upgrade comes out and you’re stuck on that version until they all update, OR you update not realizing and lose the functionality for some time.

Since I self-host a bunch of apps, it made more sense to use different apps dedicated to those features, like Miniflux or Navidrome. Not for everybody though.

It is constantly screaming for upgrades but I don't know what they are doing with those upgrades because it doesn't seem to make the core any more reliable. It's just an opportunity for the upgrade to fail and lose data, which it frequently took advantage of.

The key thing that really annoyed me is that I couldn't hardly have used it more lightly than I did, and it took about every other opportunity to fall over and lose data. I can't even imagine the plight of someone trying to run this in an office environment.

I've run it for about a year until the upgrade tipped me over. I am sure that correct upgrades (n to n+1) are fine - I tried n to n+2 and instead of explaining to me that this is not possible, Nextcloud explained to me that this is not possible and fucked up big my install. I had to recover from a backup.

    if(target_version > current_version + 1) // not allowed

It might be creating 100k hours of waste on the sysadmin side to save 30m of work on the dev side. I just don't get it.

The proper way would be to abort the upgrade upon discovering that I am going for an unsupported way.

This is not serious software.

If you don't support upgrading, don't expose that in the UI. There's no excuse for having UI that will let the user do something unsupported and then screw up their data.

Also, what do iOS people do to make their auto upload folders upload automatically without having to open the app?

I launched the container for the new version, but got an error: "Exception: Updates between multiple major versions and downgrades are unsupported."

I then tried launching the container for version 18, which should be supported, which initially seemed to succeed! However, I was greeted by "Internal Server Error" in the web UI, since apparently instead of just quitting, the initial upgrade had ruined the DB schema.

After restoring version 17 from backup, migrating to 18 and onwards, then trying to register my local apps with it, I then got CSP errors in the console when signing in. After manually editing the configuration file (which you seemingly can't do through environment variables) that seemed to be resolved.

Then, I tried syncing my local files but realized that the size of all of them is suddenly 0 KB. The files would show up in the web UI, but the size would be 0 KB. I had to rename the files a few times, which eventually fixed that, no matter how odd.

Finally, I realized that trying to open any of the pages in the web UI would take around 3-10 seconds. It seemed that there was a bug in the Talk application, disabling which then gave me the expected performance, even if it affecting how the Files application works also seems odd.

I like the utility Nextcloud provides me, but despise updating it. At this point I just stick to the core apps and wipe the whole thing when I need to update to something newer, manually carrying over the files I want to sync in the new instance.

Oh, also, out of the box the DB file locking configuration causes large file uploads (a few GB) to just hang, in a way where I cannot delete them. I had to disable locking altogether.

Honestly, it's just weird. If you need 1000 SQL migrations to go from version 1 of your software to version 20, just include all of them and don't mess around with disallowing migrating across multiple major versions. Same for any changes that might impact the persisted files on the FS. Just have updates be one continuous, long stream of patches, applied sequentially. If you do that and have sane defaults, you're good.

To not be overly negative: it's good that Nextcloud exists, is free, has a somewhat nice ecosystem around it and both the desktop apps and the phone apps for iOS/Android are okay.

I can't speak to the other points you made because this has always stopped me from investing in it. What I need in my stack is a focused Google Docs alternative, but every time I've installed it (3-4 times at this point) I'm quickly overwhelmed by the quantity of stuff it includes and by how complicated actually setting up the Docs replacement was (at the time the recommendation was to install Collabora and link it up with Nextcloud, which I never could get working).

What I did see out of the box was a slow and bloated web portal, a bad calendar, a bad video conferencing app, a file backup solution, and a terrifying app store filled with add-ons that may or may not be maintained. Oh, and now it looks like they have an AI assistant?

I'm honestly relieved to see this here and know it's not just me. I ended up going with Seafile for file backups and have been very happy with it, though I'd still love to find a focused collaboration tool I can run.

Try Etherpad (https://etherpad.org/)

> I ended up going with Seafile for file backups and have been very happy with it

I went through probably dozens of solutions and ended up with Syncthing for synchronization of data (it requires understanding how it works but once you do (usually an eureka kind of moment) this is a very powerful system.

For me using synchronization software (Seafile, Syncthing, ...) to do backups is dangerous. Borg is a very good solution (or Restik, or Kopia)

Yep. Just synchronisation on its own for backup is at best better than nothing, a first step. It fails to protect from at least three very common occurrences you need backups for: accidental deletion, incorrect update, or corruption - the broken data is quickly synced everywhere. Adding snapshots and regular integrity checks is essential.

Seafile has file version history, so it's more than pure syncing and has so far been sufficient for backups, but I also run it on site so it's not safe in that sense and the data matters enough to be worth redundancy.

For collaborative editing LibreOffice Writer has a collaboration feature. Otherwise I’ll send a copy via message service of some kind (email, txt, chat, whatever) or download a copy to my Documents folder if someone sends me a google docs link or whatever.

On that topic, it’s amusing to see when the “are we a product or a platform?” confusion jumps from its native host (publicly listed tech corps) to FOSS projects – different species altogether.

The idea of owning your own server just like any other device like laptops, smartphones and tablets, is just wonderful. But that means Linux! And Linux is absolutely infamous for poor application distribution (well technically the Linux distros). So now people make their own half-assed “app stores”, which is infinitely worse.

We would need a platform spec so that application developers can publish, users can discover and deploy, and platform providers can implement against a shared spec. That would be such a major step away from the surveillance economy and towards data sovereignty.

That’s right. And for UI, web is the obvious choice. But it’s not an application platform alone. You need at the very least file systems and networking/routing over https and maybe lower level protocol support too, that apps can use/discover without having to have a phd in Linux configuration. But to be truly useful, you need a lot more stuff wrt identity & auth, server-to-server networking, service discovery, cross-app communication, etc. Something more like Dokku probably.

> Or stuff like Helm, when you're talking about larger, more complex deployments?

Well fortunately you don’t need to open the Pandora’s box of consensus problems for a personal node (again like any other device). Also too much Yaml to make it sane probably?

That's been tried with Linux Standard base, but it seems 'standardise on glibc 2.31 and Qt 3' is not a viable way to describe system dependencies.

Such a spec also doesn't exist for Windows or macOS systems.

If you target 'living' operating systems you need living applications that are updated to keep working. We're well past the point you can assume some 20 year old binary will work unless you're running it in an emulator.

In a way, I have the same feeling. I really wanted it to work, but the install ( in my case via unraid ) broke multiple times after I started doing anything outside the basic functionality.

Eventually, I gave up on it and moved to dedicated apps for each portion I wanted and I have not looked back.

edit: syncthing for file sync wps-office for general doc work photoprism for pic library

Updates have never been an issue. And honestly I am always a bit surprised by that. I don’t update to a new version right away but when I’m ready, I change the version in the docker compose, pull, re-up the container. It performs the database migration and brings up everything. Never had an issue after using it for years. Not sure what your exact setup is, but it’s certainly not a nightmare to use.

There's like dozen scenarios where you end up with 0 byte files with no warning, error, etc. Sometimes it's an uploading issue, sometimes it's a syncing issue, etc. And they are still open, mutli-year, issues.

I ran into a version of it where existing uploads were replaced with zero byte versions, so I had to manually run a script to find them and I restore a pervious version. If it tried to do other things, like photos, document editing, collab, etc and did those poorly that would be fine. Failing at the most central task of just storing the data is not. It's still alpha quality, and if one intends to use it, it's critical you setup monitoring infrastructure to ensure file integrity (backups should go without saying).

Running it on k8s has similar issues. I only used it as sync for my obsidian notes with occasional (maybe once a month) access to those notes via web ui. I wasn’t happy with how it would reformat my notes on save.

Sync thing supports encrypted storage in VPS and is trivial to run in k8s and it’s been much better for my needs.

Yep. And any extra apps beyond the default just make upgrades go sideways. I've given up on it. Using syncthing instead (just for file syncing) and haven't looked back. It's not my favorite either, but just because it's a pain to configure. Once configured, it's been rock solid.

After giving up on Nextcloud I tried syncthing too - hated it - and most of that time was pandemic and not like I was syncing outside of my own home network.

Went back to dropbox instead (just for file syncing) and haven't looked back.

Wait until you see the "solution" for dealing with the client mangling timestamps [1]. After you've been burned by server updates, and then burned by client updates, with solutions that push the problems onto the system admin, you start to wonder if using it in the first place was a mistake.

IIRC, part of the reason they forked was because OwnCloud wasn't moving fast enough. My experience makes me think OwnCloud might have been right.

1. https://github.com/nextcloud/desktop/wiki/How-to-fix-the-err...

Look, it's not without its quirks, but it's pretty reliable. Speed and responsiveness can be further upped by using an in-memory cache like Redis/Valkey.

Nextcloud can try to be everything it wants, but as users we can ignore that and stick to file-sharing/online/cloud storage.

If one is using it primarily for file storage/sharing, it works reasonably well. We have ~1TB of Nextcloud data on an NFS share (itself snapshotted ZFS), backed up to B2 regularly.

If something happens to a Nextcloud server (and it has), restoring the data is as simple as recreating the users and copying over the data directory for each user (provided you're not using it for calendaring and such). A quick OCC command to re-index user data and we're back.

We haven't had any issues running the docker image (orchestrated via HCP Nomad).

We have also helped a friend run Nextcloud for their small non-profit, as a snap on an EC2 instance in AWS. It keeps itself up-to-date and has been pretty-much problem free for years. EBS Snapshots cover backups there.

Last week, I noticed that one of my contacts had an invalid date for their birthday. Try as I may, I cannot get Nextcloud to recognize 1/5/2000 as a valid date.

(And no, it's not a localization problem.)

I handle several nextcloud instances, one of them updated from owncloud, I'm very happy, it does a lot of stuff well and when you need specialized software you can because it's just files. Updates have been painless.

Did you read the manual? It is recommended for such size to do the first sync using rsync (or similar) followed by the CLI command to build the files database.

I love self-hosting but file storage is one thing that I don't want to risk. I've been paying for Hetzner Storage Share [0] happily to save myself the headache.

If you want a different hosting provider, Nextcloud now has Simple Signup program[1] which helps new users to sign up for a free plan with a provider near them, offering ≥ 2GB of storage. You can also browse the entire list if you want to pick one manually.[2]

[0] https://www.hetzner.com/storage/storage-share/

[1] https://nextcloud.com/sign-up/

[2] https://nextcloud.com/providers/#:~:text=Providers%20for%20h...

Assuming you start with NX11, which has 1TB storage, and before hitting the limit you want to upgrade to the 5TB storage, NX21.

Can you just call Hetzner and tell them they should upgrade your NX11 plan to NX21 in-place, or will you have to order NX21 and then move all the data over to the new instance yourself?

  Scalability
  Stay flexible with your Storage Share. Regardless of how your requirements change over time, you can upgrade or downgrade your Storage Share in a few quick steps and without worrying about data loss. Simply switch between the Storage Share package size you need by going to your account on the konsoleH, and then to "Account type".

It's fully FOSS. https://www.collaboraonline.com/code

> Due to performance reasons, we cannot support the built-in version of Collabora. So if you still want to use Collabora, you will need to provide your own server. You could use, for example, one of our unmanaged dedicated root servers or a Hetzner Cloud server. You can activate Collabora via the App Store, but you will need to use the other server for data processing. You as the customer are responsible for configuring this server yourself in the app's settings.

[0] https://docs.hetzner.com/konsoleh/storage-share/faq/addition...

https://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=nextcloud

A common misconception IMO is that running and owning your own infrastructure is somehow more secure. To that I lol, and I’m confident that the thousands of AWS/GCP/Azure/iCloud security engineers are all doing a more thorough job than you can. At the very very least they receive embargoed bugs which they often mitigate before the general public.

Also, I lol at most CVEs. Butterfly farted outside, oh uh.

Take the top one: In Nextcloud Desktop Client 3.13.1 through 3.13.3 on Linux, synchronized files (between the server and client) may become world writable or world readable. This is fixed in 3.13.4.

You mean to tell me a few minor point releases imitated umask, making world-readable [and possibly added writable]? Oh no! The tragedy! Keep in mind most clients are single user systems anyway.

Judge them on their facts, there are vulns and then there are vulns. CVEs are a sign of attention on a project. No more or less.

Either take control or sell/outsource it, no skin off my teeth. I was replying to someone making the case for 'just trust Google/whoever, lol'

My point is this nears hysterical fearmongering. I'd prefer if you don't go on, but it's more for your benefit.

Stopping before I start my own rant about risk tolerance

("A code injection in Nextcloud Desktop Client for macOS allowed to load arbitrary code when starting the client with DYLD_INSERT_LIBRARIES set in the enviroment")

But for some of us it is a bit of a hobby to run our own infrastructure. And some of it only ever runs on a private network.

I rolled my own docker setup for Nextcloud a few years ago, and couldn't be happier with the outcome. It does require me to log in and update the system and setup from time to time, but that's just a good time to drink a hot bevarage and listen to podcasts in my mind.

For anyone hosting their own instance, Nextcloud offers this scan[0] of your public facing url which might come up with something worth fixing.

[0] https://scan.nextcloud.com/

Generally you use these disclosures to make directional decisions about infrastructure. The list of fixed and disclosed CVEs combined with the legacy PHP code base doesn’t really pass the security sniff test. You really wouldn’t know for sure without doing a full code audit.

Managed services also have to accept connections from the public Internet, which on-premises solutions do not.

In theory you are correct but this is like saying keeping your money under your mattress is safer than a bank.

Yes, in theory someone can still all the money at a bank but the bank is infinitely more qualified/competent to not get robbed than you would be.

If done properly cve-s don't matter that much. You create a headscale install on a pi and the headscale port and your router's ssh (key only) are the only things visible from the outside. Take any other than a home router - aka something with support. And you are done.

I think it depends on the CVEs and where they are. If it's a software vuln that requires root or some other complex prerequisites then w/e. But, if we're talking about low level problems in either the OS or network layer (e.g. firewalls, routers) then big clouds are most likely going to have that patched and rolled out more quickly IMO.

If you aren’t being specifically targeted, then you would care about low hanging fruits discovered by something like automated scanning. Not exposing your service to the internet does solve this assuming you’re confident in the stack which provides this isolation. But managing this stack and performing risk calculus here is actually where the security horse trading happens. I think most people aren’t safer managing this themselves — arguably they’re actually worse off.

I have high standards for the confidentiality of my data. I care about things like lateral movement and the massive attack surface that isolation tech to prevent such movement has. I also won’t design monitoring and alerting, ensure a patch state, or perform code audits on Nextcloud and all the isolation tech required to secure it to a comporable level of security. Because of this, I instead reason around the cost of exploitation. I want it to be higher than what I believe Nextcloud provides and I’d rather require an attacker to use an expensive 0day to extract my data off a cloud provider like Google versus a potentially cheap one against my own infra.

Anyway, I have mixed feelings. I admire the community and the support it has by many governments, its staunch Open Source basis so it's useful for an individual or a large organization. But it is building on a lot of crufty PHP, their collection of apps is very uneven and it's hard to know what works well without a lot of research, and it's going in a few directions to upgrade. AppApi in particular is on one hand very innovative, on the other going in some odd directions. I know it is successfully used by very large organizations, but without spending a lot of time with it, it's hard to get a sense of the commitment and considerations required.

I have a synology nas and I use the synology photos app for auto uploading without loss of quality to my nas. Don't forget any picture or videos, deduplicate, figure out what you can safely delete from my phone,... All of that works great and is reliable.

The synology photos app to watch your photos though, isn't very good. Neither on mobile nor the desktop nor android TV. It's bare minimum and even that has failures.

Next cloud,... It gives a weird "I don't trust it" vibe. I want my stuff unedited unaltered, unmoved from their folder and keep your metadata elsewhere, and it just give me a wrong vibe for that. This is 100% feeling and not facts.

Immich is great but high maintenance, any update is a risk that you have way more work that planned. Their upload app need works but I use the synology one. I don't feel safe having that handle the main copies of my picture, and the dev are being super clear that I shouldn't.

Photoprism has been my go to. You need to pay for any advanced features, and unless you're solo you need them for permissions only. I love it but I feel like it's still not quite as good as I would like.

Is there any major, reliable app for that I missed?

What about Piwigo ? They have a mobile client in F-Droid. I've not used it myself, but a friend is happy hosting his own photo gallery with it.

Reading between the lines I think it is closer to dual licensed. With extra conditions for the non-public source.

Looking forward to it receiving more polish, but it is workable in the interim. The public e2ee web gallery sharing feature is killer.

For people like me, who have to contend with GDPR and NIS 2 requirements on an organizational level in Germany this actually is a god-send. MS and Google don't play nice with the local law, e.g. refusing order processing agreements, although MS has recently started to move on this topic.

We replaced MS stuff (all of Sharepoint, most of PIM) with Nextcloud and MS ADS via SAML without too much of a hassle. The sync client has its nuisances (e.g. can't name Nextcloud instances) and it's getting worse for the last few versions! But it does its job, i.e. syncing, and you can always use the cloud itself.

Do you have one NC instance or do you federate (or plan to) ?

Do you use the mail app ?

Which applications do you use to replace sharepoint features ?

Do you use the business version of NC ?

> Do you have one NC instance or do you federate (or plan to) ?

We actually have three instances. Currently they are running side-by-side but federation is planned, I think, starting October. Account management is still done in ADS and will stay there for now. SAML works as intended.

> Do you use the mail app ?

Yes, we use the Nextcloud Mail app without the AI features. Most people are using desktop clients, ie. Outlook, for daily work and it is important to us to keep Nextcloud Mail and Outlook in sync. Since the recent changes by MS new Outlook, it is also on the clock, but we are not there yet.

> Which applications do you use to replace sharepoint features ?

We used Sharepoint for centralized file storage, project management, and department sites and not much else. File storage is a given. For projects we moved to Jira some time ago, so little to do here. For department sites we use the Pico CMS integration.

> Do you use the business version of NC ?

Not yet. We are well below 200 FTE, so too small ;) We are interested, however, in the compliance and GDPR certification and we have already contacted their sales. But I don't know what became of it.

EDIT: formatting

My criteria is "when I want to show my vacation to X on the TV at a dinner 6 to 12 months after setup is it reliable or does it always need some tweak or whatever and ruin the moment".

I dont know the apple ecosystem, for me Google photos is king but I want to own and host my data, Photoprism is my current goto.

You can put it as a front application and wait (if corporate hasn’t forced a screen turning off timeout)

Not affiliated with the product.

That took some time for the dev to create and would take me at least an hour of my time to reproduce, for which im happy to exchange $5.

I could probably set up an ST instance inside iSH but it won’t work as smoothly, or create a native port myself, but I would rather not and feel thus is a fair exchange.

Of course, I would welcome and am waiting for the release of your alternative free version. Until this time I’ll take your comment as yet another “i hate apple” post.

Judging by your comment I suppose you don’t use generic drugs or any products based on expired patents not by the original patent inventor.

Sadly Apple requires doxxing yourself to publish apps, which IMO is a human rights violation. You also need to provide a working phone number to be able to install even free apps on a phone or tablet you have already purchased. These policies stand directly in opposition to their stated value that “privacy is a human right”.

The time of forced side loading support cannot come soon enough.

> Of course, I would welcome and am waiting for the release of your alternative free version. Until this time I’ll take your comment as yet another “i hate apple” post.

In my home a few minutes ago I just re-hung a framed, signed, and numbered print of some of Susan Kare’s pixel art for the original Macintosh System 1.0. It’s a common misconception that I hate Apple, but nothing could be further from the truth. Of course, you are free to continue believing incorrect things. :)

FWIW, I paid the $5 and have Möbius Sync Pro installed on my phone. But, as a Syncthing contributor, fuck them.

(Separately, it doesn’t even work right, whilst iCloud does, because of Apple’s anticompetitive bias against third party apps replicating OS functionality.)

> Judging by your comment I suppose you don’t use generic drugs or any products based on expired patents not by the original patent inventor.

I don’t believe in the concept of intellectual property. You misunderstand where I’m coming from. It’s a dick move to profit directly off the misfortune or hard work of others, which is precisely what the Möbius Sync authors are doing. It has nothing to do with patents or copyrights or what is or is not legally permitted to be done with the code.

You will note that I release all of my own software into the public domain, not under copyleft licenses. People should be free to be a dick, just as I’m free to point it out.

Ill be waiting for your public domain repo with ios syncthing build instructions - no need to doxx yourself. ill happily build it for myself - i don’t have a dev account either.

FWIW that app works much better than icloud has ever worked for me. eg with icloud you never know when the device decides to remove a file from local cache until youre offline and need to use it.

It's not because of the forced doxxing, syncthing has a foundation with a legal address and real people, using that info on the app store is a non-issue. It's not because of the $99 fee, syncthing has some cash flow.

It's been 10 years already, so why?

App Developers usually don't give a flying cluck about anything other than their apps. Battery, CPU, mobile data? Zero consideration.

But then of course the fault lies on Apple (and MS - it's the same thing in Windows - see the CrowdStrike fiasco)

But nobody wants to sysadmin their phone with rare exceptions

I like nextcloud, and always have an instance for file sharing with people. For basic features, it works good.

That said, I think it would have been better if they could focus on core features, remove code complexity, make it faster, remove bugs around the edges, make sure it just works and upgrades reliably. It could not possibly do such diverse range of applications, and now AI.

If we can advance portable specs so different open source projects can interop... that would be better time spent.

it certainly seems much more stable so far. The lack of features might be a feature itself.

The NextCloud site makes it all look great and probably easier than some parts of my current DIY strategy, but reading here I think I'll stick to said strategy.

I hope some people find it valuable though.

I definitely agree NC's core functionality could be more polished though. Performance issues don't bother me as I only host for myself but there are UI issues and bugs that get in the way of functionality. I feel like everything is 90% there but there seems to be little interest in making it to 100%. The collaboration stuff they are mostly focused on is useless to me but I can understand why it's their priority, commercially speaking.

Occasionally I toy with the idea of replacing it with separate tools for file sync, photos, calendars, contacts, notes, etc, but it's very convenient to have it all as one.

But I definitely agree that all the other things are half baked. The media player in the iOS app is extremely buggy. There's jank all over the place. The photos UI is pretty much unusable if you have a lot of photos.

And then there are the constant updates; I get notifications all the time that Calendar or some other app has received an update. I wish they'd coordinate releases of their official apps so that I can get a notification every now and then that I need to update all the apps, rather than constant individual app update notifications.

In short, as you say, everything is 90% the way there.

But honestly, I should probably just use Seafile and get another, more lightweight CalDAV server.

It totally does not work. If you try to send or synchronize just a few files it is mostly ok. But when trying to upload more than a few dozen files, things starts to not work well, like transfer being stuck, or the transfer completing ok but then only a portion of files were transferred.

That instability is the main reason preventing me to use it daily as in the end I will just use it on the computer where the main value would be on mobile.

Also sharing has been very confusing, people keep thinking they've shared files they haven't.

Maybe the problem is me / us, idk. Or maybe it's better as a "personal" cloud than a shared one. I have, overall, not enjoyed using it. I wish management would just let us store docs in a git repo, which would be much more intuitive for most of us.

I experienced the bug you mentioned about 5 years ago, I wonder if they ever fixed it.

Not using NextCloud for anything else yet though

There is some under documented good little tools that can sync between devices, browsers and Nextcloud :

- Qownnotes for notes, selecting text from web pages and bookmarking (I'm curently testing it).

- Someone told me that Zotero can do that too.

- Vdirsyncer for synchronizing calendars and addressbook between NC and uncompatible apps (Odoo mainly).

I want to love it, but it's just so not there yet -- I worry that its lack of focus will prevent it from ever getting there.

It's easy today because current IT evolution is totally skewed, derailed for commercial reasons, but that's untenable. Who think it's normal being even unable to access already made documents only because "the remote service or the network is down"?

Considering the continued shrieks of pain coming from the graphics design people this is not the case. Those people are still tied to desktop applications and look what freedom of choice they have. Virtually none.

I think you are not alone, but I also wouldn't be surprised if this frog is boiled so slow that our opinion eventually becomes irrelevant: today's teenagers are *less* proficient on a keyboard than their elders, because they no longer "type" but "tap" instead. When the average exposure to computing is reduced to dumbed-down "consumption-only" devices running iOS/Android, and that goes for a generation or more, the perception of us would likely be that of old folks yelling at clouds.

- most UIs actually have switched from the "commercial menu" model to a "search&narrow" one, now are kind of very rudimentary CLIs;

- from "commercial widget-based GUIs" back to some kind of DocUIs, today they are web, back than PostScript and alike, but the idea is essentially the same;

- NotebookUIs (like Jupyter, IDEs like R-Studio where some "GUI elements exists mixed with text) are more and more common and well, tomorrow someone will rediscover Emacs buffers ideas...

So, while most computer users are consumer only, those who can't consume only a little step at a time, following a kind-of Greenspun's tenth rule reimplement partially and with many bugs and limits old ideas. I do not know where we could go in 10+ years, but... Even with all the bug-ridden modern overhead, we actually rediscover classic solutions, at a certain point classic desktop computing will came back forcibly IMVHO...What I do not know it's how much time and injuries we will suffer in between...

I just plain can't use the thing. The only thing that works well is automatically uploading new photos from my phone. That's it, that's the only thing.

My server is old, yes, but we're still talking about dual 8c/16t Xeons, 128GB RAM, and a 7-disk striped RAID storage. It's no slouch.

I can connect to it with a bleeding edge gaming computer with the latest i9 16 core whatever and a fancy GPU with any browser direct over gigabit Ethernet and it still, somehow, takes a full thirty seconds to show the login page. Clicking on anything in the browser is at minimum a 15 second delay. Trying to browse files takes a good 30 seconds just to switch views to the new folder and another 30-60 to render a plain text list of files. Browsing photos is only slightly slower. I get the exact same level of performance when my client is an ancient first generation i3. It's astonishing, really.

I've debugged this far beyond my abilities and the best I've been able to determine is that the browser is spending most of its time just waiting for the server to respond. The server shows no interesting trends in CPU usage, so I have no clue. I can only assume it's attempting to render server-side on a single thread. Given the poor single thread performance on the Xeon this seems fairly plausible.

But regardless, this is by far the single worst performing website I've ever seen anywhere, ever. We've all seen horrendous single-word-per-page scroll-scroll-scroll slideshow websites that take ages to load. Nextcloud makes those sites look like bare HTML GeoCities pages.

It's atrocious, unacceptable, and unusable. There's no excuse for this.

And yes, I'm using the AIO docker image. I've also tried bare metal installs with the same performance. I've tried a more modern i5 machine. I've eliminated any problems that could be on my part and I'm left with the conclusion that nextcloud is just inexcusably bad.

Which is a damn shame because this is everything I want from self-hosting. I'd love to get my family onto this. Have a central place to share photos and files. Keep everyone's devices backed up in a place I know is safe and secure. But I wouldn't inflict this thing on my worst enemy.

This all sounds like hyperbole, but it truly is the worst of the worst. I have never, since the invention of the internet, seen a website this bad.

Edit: oh and also all of the desktop clients are barely-working garbage. The UX feels like it was designed by aliens who heard about human interfaces in a Chinese-room scenario. Utterly absurd nonstandard unintuitive garbage. The android client isn't much better and the iPhone client is famously broken on purpose.

There is obviously something wrong in your NC setup.

I've tried everything from hardware to network configs, always the same abysmal performance

I have only just started using it (and hence wanted to check what others in HN were using it for). Seems like the personal use case for file syncing is difficult; I'm more interested in self-hosting Kanban and other things I don't want to pay yet-another-SaaS a per-user-per-month for the small business tax. Setting up NextCloud with an IdP wasn't too bad, a little weird on Googe's side that it doesn't permit SAML SLO (but that's my not-terribly-deeply-informed opinion).