I'm sad not to be able to recommend Cloudron as the best open source paas since the license change.
It has in effect changed my relation to the project, from an invested advocate to a simple client.
I totally agree with this part. More than that, I would never have picked up Cloudron at all at the beginning if it weren't open source.
And as to contributions, I am the author of one of these rare contributions ^^ (to make the platform compatible with the OpenVPN app), and I would definitely not have contributed if it were not open source.
TLDR: I am 100% in favor of switching back to an open source licence.
(As for the precise licence, I do not really care, be it MIT, Apache, GPL, AGPL ... whatever.)
Given the nature of the responsibility of the Cloudron system, security is the biggest aspiration for me from open-source, along with a security reporting process that allows for private communication of any issues found.
I recommend including this somewhere on your site:
My page own open-source on our own platform (WP&Woo stack) for interest:
Although we haven't open-sourced that whole stack yet, it's planned.
As for security issues to have a private conversation, please see https://cloudron.io/security.html
Also as mentioned earlier, we do share the view that it is useful to introspect the code to see what is happening on your server, this is already achieved by our source-available policy, so feel free to audit that in the git repo.
As could probably be gleaned from my previous posts, I’d love to contribute to the code. Adding an option for disabling unsafe backup notifications, adding support for inter-app network communication (so my OpenVPN Client Cloudron app I’m building’s network can be used by any Cloudron app with a quick restart of the app), adding multiple domains for Wordpress Multisite, etc.
I do feel a little bound to the main developers roadmap when I’m loving Cloudron more and more with each update. Some fixes are just mere nice-to-have (the option to disable unsafe backup notifications), but some are necessary features I need. I’m no stranger to building what I need for myself and then letting others benefit from it. But a pull request for these features is impossible if it’s now closed / proprietary. Which, by the way, I totally understand. I just wish I could contribute to the project my main features so the developers don’t have to do so (eventually) and I can get what I need as soon as I need it.
I honestly didn’t know there was a GIT somewhere for Cloudron so I can at least start browsing the code to get familiar with it.
Everyone contributing in the forums, codebase and apps are heroes!
I wish I could get more involved on that side but the next best I can do is keep telling every developer I work with and know online about it to try it, and get you more developer users — because I think we all can see the value in both the platform and this community.
I shudder to think of doing all that Cloudron does any other way now, and have been through pretty much every way of doing Sys Admin in 20+ years of tech.
@Lonk Nice. I work with a team of 10 devs, mostly WP & Woo but inevitably the full stack & dev-ops. They have a good 1,000+ ticket backlog from me on feature development but I've been introducing Cloudron for peripherals apps for now, like Bitwarden, PrivateBin, NextCloud, Email etc.
We'll definitely be getting more involved, the incentive I always try to work with is that all our team owns the codebase, so everyone benefits from the collective. I always try to make what we do portable, so anyone could fork & run with it at any time but the greater incentive to collaborate is in the experience of the team.
As a separate project I'm looking at starting an open-source tech fund looking to invest in things like this directly or indirectly with sponsored development. Just matching investors to their tech appetites, returns expectations, and overall business models we have as users of the stuff we develop among our team and with other communities.
Recommend following Sahil, on Twitter the creator of Gumroad, he has lots of experience and commentary on this subject.
Just a thought for the business-model side of open-source because as much as I love and promote open-source, I always look at what the business model is behind it. Any software choice is a long-term commitment and I want to know that progress is motivated and sustainable.
Partner programs / subscription levels. Odoo has a good example of this. The return on investment for the Partners being implementation referrals.
Hosts referral revenue share
Many of the FOSS apps included with Cloudron use this model.
As I say, just thoughts, with the utmost respect for all that the team here do and it's their work to do with as they see best and whatever works for the schedules they have and generously share with us.
The replies seem to be coming more from developers lately, people who have skills enough to contribute, who also hold strong beliefs about open source. This is good, and by and large it is pleasant to see an overall positive approach to Cloudron.
I am a non-developer, but am also someone who likes the idea and principle of open source. I want to get away from proprietary apps and data lock-in. From my perspective, Cloudron has been an incredible way for me to learn about and access a plethora of open source apps, the delivery of which just works! I hope the various principled devs out there can appreciate a users perspective like mine, and just how radical Cloudron is. I've tried other open source attempts, and frankly they are all lacking in deliverability; checking out their forums reveals not even the devs can figure out what went wrong when something does; and rarely it seems are the project heads in the forums, like the Cloudron Team is, since the resolution of any problem relies on the upstream devs pushing fixes that may or may not come in time. What a mess for a user like myself, and I think there are many many users like myself. I think we are one mid-layer that will help make open source more understandable, reachable, and friendly, to all our non-techy families and friends, they who think only of Skype, Word, Facebook, but themselves are increasingly becoming disgruntled. They will never read certain /r/ nor news.ycombinator.com articles, but they will listen to me when I set up an open source chat instance, using a vanity domain, with email, on Cloudron, and it just works!
Another positive aspect of Cloudron's approach is that they don't try to cloak the open sources apps they offer as though they are something the Cloudron team created. I've come across a few such endeavours, and it worries me when I think there are regions and countries that think ****blog, *****forms, *****pad are all made by *****soft! It verges on deception, even though, after some digging, a user can find out that the software is actually made by other open source efforts. I hope this means that as more people hear about Mastodon, or Element, or booktype, etc., and look for options, they will be led to Cloudron. I for one came because I couldn't for the life of me get Rocket.Chat, Taiga and SOGo all running on the same server, by hand. Not a problem for Cloudron!!
Keep it up!!
Love your testimony. I am a developer-user (well, just developer for now - but I could see me running this for production) and I'll 100% commit myself to the Cloudron ecosystem as long as the current closed-source sections (like Dashboard access) stays un-obfuscated post installation. So far, neither have the developers have spoken into if that was intentional (not obfscating the code). But, I hope it was.
Interesting weekend read: https://plausible.io/blog/open-source-funding
Didn't think that was the best article to be honest.
It mentions charging for support, but not the good ol' charge for support + updates model which has proven successful for loads of open source projects, including the first ever billion dollar open source company Red Hat. But also think of the vast majority of the WordPress premium plugins and themes ecosystem.
EDIT: I just realised I could do an Advanced search for "most of the packaging work" in Announcements to find lots of them (although not all, as e.g. that didn't find Moodle).
In no particular order:
- @atrilahiji did much of the work for the Moodle
- @doodlemania2 did most of the packaging work for Pixelfed and Apache Guacamole
- @jimcavoli did most of the packaging work for Metabase, Grafana, Snipe-IT and Grav CMS
- @thetomester13 did most of the packaging work for Firefly III and PrivateBin
- @fbartels did most of the packagingwork for Statping
- @ultraviolet did most of the packaging work for Vault, Trilium Notes and Apache Guacamole
- @msbt did most of the packaging work for TeamSpeak, YOURLS, Alltube Download, Bookstack and helped with Matrix/Riot
- @syn did most of the packaging work for Mastodon
- @Felix and @iamthefij did much of the heavy lifting for Bitwarden
- @murgero did the initial ground work for Directus
- @cve-random did the majority of the work for Jellyfin with the help of @mehdi
- @mehdi also implemented OpenVPN
- @sumacinitiative helped out with LimeSurvey and SearX
- @BrutalBirdie did most of the work for Greenlight
- @erics packaged dolibarr
I did a few other searches too and think I've likely got most of them now, but who is still missing?
Hey everyone tagged above.
First of all: THANK YOU!
Second of all: as people who've actually contributed apps to Cloudron, I'd really appreciate your input in this thread
Hey @marcusquinn @yusf @murgero @d19dotca @fbartels @will @necrevistonnezr @mehdi @msbt @Lonk @doodlemania2 @imc67 @ruihildt @JOduMonT @atrilahiji @scooke @heliostatic @jimcavoli @Hillside502 @robi @thetomester13
As valued contributors to this forum, if you've not chimed in here already, please do so!
To all those on both lists above: HUGE RESPECT (and I really want to hear from you)
@jdaviescoates Woop - jeez - didn't know I could get stars for being so distracting
I'm in the good idea but not in a hurry camp - FOSS comes with admin work - and for now, I can see development progress is pretty healthy.
I've equally worked on a stack that should and will eventually be OS - but when I can commit the time to supporting the community expectations on top of our own needs, so I see both sides.
It may only a matter of time - but for now, in source available I think we all trust. So yes, but patiently