The trouble is, they confirmed that there are differences, and there may be more going forward. I do think itās significant that there is now a different proprietary build, however great or small that difference is claimed to be. It seems to me that potential customers are led to believe that the difference is important, while the community is assured that the differences are trivial.
When you say that thereās nothing we can do, and we should just focus on the features that are permitted to the community, you may be right. But, we are then dealing with another self-hosted cloud software with a community version and a proprietary version, and there are several of those. Owncloud itself only restricts the features it deems to be for enterprise needs, and many of the arguments now made here would apply over there. If Nextcloud is to move in that direction, then it loses that (in my mind) huge advantage of being 100% open source, and I think we ought to be clear about that.
Just to be clear, we still support Collabora and work with it. It is just technically much, MUCH harder to integrate Collabora the same way as we now did with ONLYOFFICE. And believe me, everybody we talked to already told us that what we wanted (make one of these two office solutions just work out of the box without docker or reverse proxy stuff) was IMPOSSIBLE. Seriously, everybody said that, including the people at Collabora and ONLYOFFICE. And it did take a year of work, that is a significant investment for a small company like ours. Especially for something that makes us no money, and never will, because it is purely for home users. Again, I assume business users have the people, time and expertise to run a single docker container and set up a reverse proxyā¦ but a private user like myself does not, and for a private user this rocks.
But we would love to make this work with Collabora, too, believe me. We will put in effort, if we can, and weāve talked to them already about it - multiple times even. And weāre not done talking, so stay tuned
(and yes, they were our first choice. But technically, ONLYOFFICE was doable, Collabora really impossible. Maybe we can still change thatā¦)
Jos ! I hope you see the last post about that OnlyOffice donāt support anymore editing files from the Mobile NextCloud.
It disgusts me a lot.
You work a lot for this.
It worked
You advertise about this feature
And now they took it back form the standard users.
Do you have plan about that (sorry that off topic but we can continue here OnlyOffice removed mobile editing from version 5.5.0 of Community Document Server )
So you conflate a few things, which resulted in some grumpy replies from myself. Sorry for that.
First, open core in ONLYOFFICE. When we started working on the integration, the two releases (the community and enterprise version) were 100% the same, except for the 20 user limit in the build for the community. But their last release, which came after we released Hub, DOES have differences. And yes, that is open core. And I am not a fan of that.
But we canāt control that, they have the copyright and the right to do that. It does not make their community edition worse, and if they would make it worse anyone could submit patches and make it better - that community edition is still protected by the open source license. We would certainly look to see if we can integrate another office suite (see my reply before) and as community we can of course also consider a fork. Not sure if we, as in Nextcloud GmbH, would be happy to fork ONLYOFFICE - they are a partner, and friends, and it would be a LOT of work just to keep it working, let alone improve it. Any idea how much work it is to maintain an office suite? It is INSANE. But in principle, I agree with you it isnāt great that there is now a feature difference between enterprise and private version.
Then again, we didnāt know this was coming, we didnāt plan for it and we still deliver now a great office suite, the fact there there is a version of it available with some more features doensāt make what we deliver any worse. And if another office suite comes out that is open source and better, maybe we can integrate that.
You note that you think it would be better to not have something than to have this open core product integrated - that I personally disagree with, but I think we can agree to disagree on that one. It is a matter of pragmatism, and we at Nextcloud tend to be very pragmatic.
Second, about Nextcloud and open core - the code we build and everything else around it is not open core. That is NOT an opinon but a fact: it CAN NOT legally be open core. We donāt have the copyright on Nextcloud. Our employees keep their own copyright, so do contributors - without a CLA there is no open core. So please stop claiming that, it is nonsense. The AGPL sets the rules for how we play, and we follow those rules. That is 100% open source. The AGPL does not require us to do releases, it does not block anyone from making a special build like RHEL does with Fedora and we do with our enterprise version. The difference between open core and AGPL is that the playing field is level: YOU can make a special build of Nextcloud. And everyone else can. And everyone is allowed to do that. And you have no right to stop anyone from doing that, neither do we have that right.
That you donāt like it that some people and companies would want to make a special build which they only give to a subset of people (customers), well, sorry - but that is completely AGPL compatible. Sorry, you can disagree, but that just makes you wrong - go read on the FSF website for example, or watch the talk from myself at FOSS Backstage 2 weeks ago, I talk extensively about business models, open core and other things. Iāll put a link to that talk here as soon as I have found it
Well, the open core part relates to the Nextcloud fork of OnlyOffice. If you integrate a fork of the open part of some open core software into Nextcloud Hub (complete with the restrictions at their request), and then resell the proprietary version for those that want to lift those restrictions, I think it is quite fair to say that Nextcloud is engaging in open core for that part of the functionality ā itās just been sort of laundered through a third party.
I understand that you say this has been done as a gift for home users, but letās be honest ā it could be presented that way, but itās also a way to push people towards buying proprietary software from you. Youāve made it pretty clear that anyone needing needing more connections ought to pay up, and promoted the proprietary version.
As you do not own the copyright on the Nextcloud code, and are yourselves using it under the AGPL, are you not obliged to publish the code of the modified enterprise build you release, or is this only distributed to the customers themselves? If so, and the enterprise code is restricted to paying customers, then it seems that in practice it still the virtually the same effect as having a proprietary version (at least from the perspective of the wider community.)
I suggest you read a bit about how the GPL and such licenses work, and what open source really is. In simple terms: you are correct, only those who we give the product have to also receive the source code from us. There is no obligation for anyone to share anything on github or any other platform. This does not make anything proprietary - you really should read up on what that word meansā¦ Proprietary code is code where the recipient of a binary doesnāt get access to the source.
With open source, you do. But if I give you a binary, and the source, because you paid me for that, I am under NO obligation to also share that code somewhere else, or contribute it anywhere. The vast majority of code written is special-purpose, consulting projects done by various open source consulting firms for specific customers. Most of them never contribute this code back to the upstream projects. And that is 100% legal, even though it is a bit sad sometimes. (I say sometimes, because often these features are really not relevant for most users, or just terribly badly writen)
We also do lots of consulting and only put the results back in Nextcloud if we think it is useful for most users. But our customers often have features and things you will never see and wouldnāt care for. That is true for any open source product, and entirely normal.
You are really getting very upset at a company that shares about 90% of the code we write, while 90% of the code written by other open source companies is never contributed back to the upstream project. In case of Nextcloud, there is a HUGE list of companies offering support, hosting and yes, developing custom apps and features that you will never see because they donāt want to share it. Perhaps focus on those?
Similarly, you are angry at ONLYOFFICE for not giving everything away as open source, but they are under no obligation to do that, they donāt even really get any community contributions - and the vast majority of companies does NOTHING open source. I would suggest you direct your anger at the proprietary companies instead.
Edit: let me add a bit more detail here, so you see how other companies do things. Collabora, which you mention as a much more open source alternative to ONLYOFFICE, does this:
They take LibreOffice, they add lots of improvements, some from the offical branch some custom. Then they make custom builds, some for specific customers with specific features, others just āgenericā for all their customers. Only customers get those.
And they contribute most of this back upstream. Not all, as some of it is customer specific.
ONLYOFFICE could do the exact thing, and it would have the same effect - an enterprise edition with features you canāt get your hand on. Just like RHEL, by the way, Just for customers who pay.
Sure, those customers have the source. But you donāt!
So, you see, āopen sourceā doesnāt mean āeveryone gets everything for freeā. No, thatās mostly Nextcloud that does that. Almost all other open source companies do FAR more special things for customers. Way more than just a āthemeā in their enterprise version, like we do, and an occasional extra bugfix a week or two earlier than the rest gets it.
So you see, perhaps, why Iām a little frustrated youāre seemingly so angry at us for working with partners who are less āpureā than we areā¦ Because, really, it would be near impossible to find a partner for us who does what we do - give almost everything away for free. Collabora sure doesnāt, neither does ONLYOFFICE, indeed. Most companies who would give everything away like we do would simply go bankrupt. Collabora and ONLYOFFICE sure think they would if they did - and I canāt look in their financial books but they might very well be right. Doesnāt make them any less āopen sourceā.
Maybe we should ask you (not really you Jos, but someone at Nextcloud that could answer) How do you get money for making Nextcloud inc working ?
Does the partnership with Collabora and then Onlyoffice gave you some ā¬ ? Or it was just sharing software and advertising ?
What is the % in your financial that Nextcloud get from Support ?
I may want to create a new topic for finding more ways to get ā¬ to Nextcloud.
Maybe from donating or a crown funding asking money to developp a feature that users wants but who need investment.
I donāt really need to pay for a support from Nextcloud, but i will be very thankful for paying you every month (like patreon) a % from the ā¬ i get from your software. With nothing in return or maybe taking a little consideration about our problems or questions.
OnlyOffice do the job quite well, itās effective and without a lot of ressources consuming for the server. The 20 users limit a ~quite~ fair. But as his way of opencore, we could be Ā« betrayed Ā» like today with the removal of a feature form the community version. I went to the Community version github, if nobody help to develop this software is because the dev donāt work with the non-dev. When i open a case in Github for Nextcloud server or calendar app, i get answers from non-nextcloud dev and from nextcloud dev, itās really nice . One of my issue (amelioration) get granted and merged into the calendar app. This is how github should work with every piece of software.
To be short maybe a grid with : you want this feature ? So give money or give time to dev could help nextcloud to be more closer to his users.
I can absolutely live with that regarding a great software I am using for free, where I have the possibility to interact with the devs and propose changes that sometimes even make their way to the product
But could you please just give an official statement that security fixes will never be delayed for the community? This question has arisen and it remains unanswered.
Great, that you invest a lot of time in answering here in the forum and setting things straight
Yes I understand that, thanks, and what proprietary means. The part I was not clear on is that Nextcloud is itself an AGPL user of its own code, and exactly what that would require in these circumstances. I originally thought Nextcloud was in a position to release a proprietary Enterprise version - if thatās not the case, then I agree itās an inaccurate word. A private paid version is a better way to put it.
But I think you misunderstand what I was saying in my last post ā that if you sell a private Enterprise version (even if those paying customers get it under AGPL), from the perspective of the community it has much the same effect as projects that have one version for the community, and a different version for paid users. It seemed very clear, in the early days, that the intention was for there to be one version of Nextcloud for everyone (notwithstanding custom modifications, which is a separate matter), and not one codebase for the community, and another for paid enterprise customers. I know you say that the differences are minimal (but note that it could increase going forward) yet it is now true that the code has diverged between the two, and we do have that divide.
The larger point here is not about whatās technically permissible under a given license. The question is about what sort of an open source project a given piece of software is, and to what degree it is a truly open community project. (I covered this distinction, and what Frank originally had to say about it, in this thread.) It would be possible to have an opensource product that uses undocumented obfuscated code to make it virtually useless to anyone else ā it could still technically claim to be open source, but have no credibility as a community open source project.
You say that the way some open source companies behave is āsadā. Perhaps thatās a better way to look at it ā is this going to be a sad type of open source project, or a happy type, going forward?
Nextcloud now ships a restriced fork of an open core product as part of Hub, and touts a proprietary version for people that need to go beyond the strict limits - thatās just a fact, whatever the motivations. We now have a separate Enterprise version of Nextcloud for paying customers only. When asked about the community foundation that was originally discussed as part of making Nextcloud a more community-oriented project than Owncloud, you told us maybe when you have the billions of Redhat. And anybody that compares the rhetoric of Nextcloud at its inception, with your recent rhetoric in defence of these things, will see the stark difference.
IMO, these developments lean towards the sad side of things.
No, Iām not āangryā at OnlyOffice (if you are detecting any anger itās because of the repeated attempts to imply privilege, entitlement, and expecting to get everything for free.) They are free to choose their business model, and the community is free to respond to it.
But, being open core with proprietary extensions makes it a very different type of open source project, and one that conflicts with the stated values of Nextcloud. It is surely not that hard to understand why some of us see it as problematic to integrate it. Itās very likely because of the type of company they are that you see so little community contribution. And, when Frank has stated that the open core business model is unstable and poisonous to a community, it does seem rather odd to adopt it into Nextcloud. Their github certainly looks quite poisonous today. Itās not about me insisting on some extreme standard of āpurityā - or at least not any more than Nextcloud used to.
Yet it is true that a layered model of financing, maybe similar to the Blender one (https://fund.blender.org/ ), could
become interesting enough for Nextcloud GmbH
become a start for Nextcloud GmbH to develop a strategy towards small/medium businesses
become a way for the community to be involved
Additionally, I hope that @jospoortvliet will not get mad with me if I reiterate a suggestion I gave before.
I am finding really interesting information āburiedā at post 67 of a long thread in the forum. And I do think it took Jos time to write it down.
Maybe a section in the forum with a summary of this information could be usefulā¦
Maybe some help with it from the community could be usefulā¦
Maybe some of the readers could help in developing this document / these documentsā¦
Thanks Jos for the insights, always food for thought!!!
Iāve actually written down a bit about open source business (I recently also gave a talk about it) and hope to publish it on opensource.com - kind of explaining how the subscription business of us and RH works, as opposed to what most companies do in open source, which is consulting, which is what I described above - that results in most code not becoming public and shared by everyone.
Now let me respond to a few other things
Well, not some, ALL. Your really think that more than 5% of the code written for the Linux kernel ends up in whatever gets shipped by Linus? ALL software companies that write open source donāt share much, probably most of that code. Every Android device is, as you say, āa restriced forkā, of an open source product. The same as open core, for all intends and purpose. We are one of the few that share the vast majority - yes, ONLYOFFICE and Collabora are with us in what is probably the top 1% of companies in terms of contributing the vast majority of work we do. Pretty much every company doesnāt. And we all strive, including Collabora and ONLYOFFICE, to share as much as we possibly can without killing our business and the progress we make on the product. We always have. The conversation we have here is really entirely unrealisticā¦ It is also frustrating to have our motivations so called into question.
We started a company with just one goal: make Nextcloud better, as quickly as possible, to give people as good an alternative to Office365 and Google apps as possible. We wanted to do that sharing as much code as possible. This is a super hard balance because if you share less, you make more money, which you can put in making the part that you DO share better. Every day we have customers asking us for special features that are only useful for them. Do we put our engineering time there and make less of our work available for free, or do we focus only on making the public product better instead of getting paid? There is no way to run our business without doing some of this non-public work because no big company would pay us a dime without some consulting. That German government deal - came with something like 1000 days of consulting work. thatās 4 or 5 years worth of work that most of you will never use! And it would never have signed without it. And most competitors in the deal would not have contributed a single line of code to anything open source, by the way, so that we do 1000 days of work and can only re-use maybe 100 of that as part of the open version of Nextcloud means 100 days more than any of the others that could have gotten that deal.
I know most of you here arenāt into business, of course, I get that. But the reality is that it is SUPER hard to get paid for support. Nearly impossible, actually. 99% of companies donāt want to pay for that. Their purchasing department simply refuses it. They have internal rules that say stuff like āa support contract can NEVER be more than 20% of the value of the licenseā ā you can imagine how hard it is to sell to companies like that if your license is 0 euro, by definition! And yes, this is a major reason why we now āsellā an enterprise edition. And you all make my life harder here, because we have to tell these companies it is super important to get the enterprise edition because it is what they pay for - while, well, you all read what I said about the difference and how big it isā¦ Letās just say - the marketing goal is to make it look as big as possible on our website
Maybe, if we had had a RHEL like enterprise version from day one, we would now already be able to pay 100 developers, making the Nextcloud we release way better than it is today. That is the balance we have to find, and until now weāve always been way more on the side of sharing more. We discuss this balance all the time, I literally had 2 rather fiery calls with Frank about this very subject yesterday.
And yes, if we thought we could have paid more developers by running a charitable organization, we would have started a non-profit. I just donāt think that works, at all, for what we do. Tell me how many dozen developers Krita is paying with the millions of users it has? Including rich movie studios by the dozens? Right. And they are lucky that they now earn most of their money by having a paid version in the Microsoft app store. Which means 30% of the money that Krita earns goes to Redmond. Yay (sorry, Iām still one of those old school Microsoft haters, sue me)
Yes on that
So we will never ever delay an important security fix. If thereās something like a remote execution issue, something that can lead to serious data leaks and all that, I can promise you that every person in this company, including myself, would quit if we would even seriously consider doing anything other than releasing that to the whole world as quick as we can. That is completely unacceptable - there are 300-400K Nextcloud servers on the web and we could never do that to the people whom we do all this work for - thatās still private users, remember.
Now there are sometimes smaller bugs, not important enough to do a special release for. Those would just come with the next update, but we inform customers and until now, we have often then shared a patch to them or info on how to work around or fix the issue. Those would go in our enterprise build for a few weeks. If you wonder how often this happens - a few times a year, in the past. And no, before we did share such info to customers, we didnāt share it at all - you know how security is (I hope). If we would tell people about every hackerone report before the fix is released and deployed by most users weād just be telling bad people how to steal your data! We have always been responsible with this and of course continue to do that.
So letās talk about money. The hard part here is that some people might immediately think weāre all getting rich. We sure arenāt, Nextcloud tries to pay ācompetitivelyā but I could literally earn 2x more if I would take a job at one of the big corporations that regularly offers me a job and the same is true for everyone in this company. Donāt get me wrong, weāre not poor, but a pro-tip: if you want to get rich, donāt work for a small open source company.
But yes, as our recent announcement said, weāre growing. And that money goes back into Nextcloud. Now we are, I think, one of the most successful open source companies with the model we use to earn money: support as much instead of consulting. In our case, it is about 1/3rd to half consulting and half to 2/3rd support. We say no to money all the time because the vast majority of customers is happy to pay for consulting (the special features that Semjel would probably call proprietary) but not support. If we took every dime, it would be 95% consulting and 5% support, by the way. And our company would be much bigger. But the consulting does not benefit any of you, because it doesnāt benefit Nextcloud - that code usually doesnāt go in. The support is what pays for the development of Nextcloud, as well as doing releases, answering questions on forums, fixing bugs reported by home users on github and so on.
Ok, it is not THAT bleak, that half to 1/3rd consulting - we try REAL hard to only take on consulting that DOES go into Nextcloud, or we add some extra time and effort (unpaid) to make it go into Nextcloud. So our 1/3rd - I would guess that about half of that ends up in what we release. The other half really is too specific to be useful. But if we would double the amount of consulting we did, none of that extra would go into Nextcloud because we already take all consulting work that fits with our product goals
But if youād ask the sales team members theyād tell you it is super hard to sell and most customers donāt want ājustā support, only consulting. I really donāt want to go and complain in public - Iām a marketing guy, OK, I have to be positive. So thatās why we always ARE positive in our blogs and the website. But believe me when I say it is 100x harder than any of you who havenāt tried to sell a support subscription can begin to imagine. And that is why almost every open source company that tries to build a product themselves and live off support - well, does not exist. Most try to bundle other products (notice that that is actually Red Hatās model!) or even compete (and kill) others that try to build a product by bundling that product for free, rebuilding it without restrictions, offering competing support without contributing and so on. Or they do hosting (where their hosted version of course has lots of stuff that isnāt available for free). Or they do open core and go more and more proprietary over time (THAT is the issue I have with that model, not the fact that something is open core in itself, by the way - I think open core is fine, but the pressure to go more proprietary is too big to not give in for most, esp under pressure of investors). Or they just give up and do consulting. Or they stay super small. Or just give up completely.
Give me a list of 10 companies that really do the Nextcloud model of:
really making your own product (not bundling others like distroās do, or something like Kopano does)
making it super easy to install and use for everyone
earning on support and just a tiny bit of consulting
not be venture capital paid and then go all wrong (kuch docker kuch)
As I wrote, Iāll do a post on opensource.com about the business part, and I recently had some discussions at FOSS Backstage, but the above (āI donāt want to complainā) is true there, too. Itāll be a positive story. Sorry
The fundamental problem is, if the conversation we have here is entirely unrealistic, that is because we were given this unrealistic understanding by Nextcloud. None of these āunrealisticā ideas are my invention. If you want to tell me that Nextcloud was itself naive in the rhetoric and the business model it put forth originally, and the intended direction had to be modified in order to survive, then fine ā please say so. But please donāt try to make out that the naivety (if it is such) comes from the community.
I wonāt quote again what Frank told us about the position on open core software, suffice it to say that it was in strong contrast to what youāve had to say about it here. But this is what he told us about a sustainable model of selling support:
āOk, so letās talk a little bit about the business model, because everything Iāve presented here is so nice, and free and open, right? You might wonder how do we actually make money, which is important if we want to be sustainable. We donāt want to spend money that comes from a bank, we want to spend money that comes from real customers. Have to be sustainable, so we have to have a real business model, and here we got our inspiration from the biggest open source companies out there.
ā¦
This means everything we do is open source and free software, and we sell support subscriptions to companies. And we focus here on big enterprises. There is no way that we try to monetise home users, or small installations, or whatever. This is not the plan. We sell support subscriptions to the big organisations ā the organisations with thousands of users. This is really important. The way this usually works is that if youāre the IT department of a big bank, or a big government organisation or a big university, then you usually want to have an SLA, a phone number you can call. If everything explodes, you want to be able to call the developers who wrote the code in the first place. And this is what we sell.ā
And you say things like this:
And where Frank also considered smaller non-paying organisations to be customers too, you say things like this.
To be clear: companies are not people and have no right to privacy, so no, we donāt care about companies other than as a way to earn money to achieve our goal of giving PEOPLE privacy.
So, perhaps you can see the contradictions that are frustrating from our side of things?
If Nextcloud were popping into existence today, and your mindset and arguments set the tone and position we all understood before going in ā fair enough, and some of us might have made different decisions. But, the open source world is a broad church, and, to stretch the metaphor, Nextcloud set out clearly what type of denomination it was, and thus what congregation it would appeal to. Now, a couple of years later, we seem to be getting a very different sermon.
No, I think I already clarified that for you ā I would call them customisations, and there is a big difference between bespoke work for individual clients, and providing separate general builds for paying and non-paying customers.
Right, this is also a contradiction that struck me. And I did think that either the community, or the paying customers, are being given a misleading impression, because both positions canāt be true. I appreciate your honesty here, but I think itās rather problematic in itself, in terms of what paying customers are being led to believe. It might be frustrating to you that sometimes your claims are called into question, but right there is an example of why someone would look at you more skeptically. I suspect, when you write your article on opensource.com, you wonāt be mentioning that part.
Maybe so, but the point is that back on day one, the idea of a separate Enterprise version was anathema ā to contrast with Owncloud. And I believe that Redhat actually does publish the code of their Enterprise version, which CentOS tracks very closely.
Well, that article is written by a particular tech investor with a significant stake in Redis, writing that as a justification for the controversial Redis move to a non-FOSS licence. I donāt think that constitutes strong evidence that āopen source doesnāt work, financially.ā And, again, if that is what weāre saying, that sharply contradicts the Nextcloud position on its business model from 2 years ago.
There are no contradictions in those statements. Weāre being optimistic about what we sell, but the majority of our money IS made on exactly what Frank said: support subscriptions. I merely stated that 1. it is hard to do that and 2. it took us a long time to get to this point and still we have to do quite a bit of non-support work to pay the bills. None of that contradicts what Frank said.
And sorry to burst your bubble, but if you feel we created unrealistic expectations welcome to reality. Please donāt go and sue everyone who says āif you try you can do anythingā and similar stuffā¦ Obviously we communicate what we WANT and what we try to achieve. That does not mean it is hard or that we immediately get everything perfect.
Welcome to marketing spin. We obviously portray things in a positive light and try to be optimistic as much as we can. Every human does that (or do you think the facebook page of people reflects their real life?) and I doubt youāre incapable of understanding that. I pride myself on never lying. But Nextcloud wouldnāt be where it is today if we werenāt positive and upbeat about what we are doing. Granted, most companies suck at that or take it to the point of outright lying and making shit up and itās a hard balance to strike. If weāve gone overboard in your eyes - fair enough, sorry for that. But please cut us some slack on that. Here are a few interviews with Frank where they push him a bit on these things:
No, the anathema was that customers would get a proprietary product with special features. That is what open core is - proprietary software. RHEL is a special product just for customers and from day one weāve said we follow the Red Hat model. A special build for customers was implied. The fact that we didnāt until some time ago (and that the difference is small) doesnāt change that.
RH publishes most of the source, I suppose. So do we - it is in github. We donāt plan on adding special features or things like that for customers - just the things we describe on our website. Nothing that wonāt end up on your system as community user.
(as I said above, we COULD following the license, but thatās a rabbit hole we donāt want to go down and, indeed, it would be not that different from open core. And perhaps that is your worry).
āRed Hat business modelā also includes doing a LOT more than support subscriptions. Red Hat does training, consulting and development, they have a customer portal with documentation just accessible to customers and much more. All things we do, too. Well, we donāt have certifications yet, maybe some day.
PS I wonāt be responding anymore, but you can have the last word if you like.
. One of my issue (amelioration) get granted and merged into the calendar app. This is how github should work with every piece of software.
or give time to dev could help nextcloud to be more closer to his users.
