Hi! I’m a software developer working on a suite of collaborative self-hostable apps to replace proprietary services I couldn’t find a good replacement for.
I am writing this post to seek opinions and ideas now that I am still in the early prototyping phase, before it’s too late to change track.
My idea is to develop a collection of simple single-purpose apps that do one and only one thing. The first app will be called Simple Notes (mirror), a replacement for Google Keep. Every operation is encrypted locally on the client, and the server never sees plaintext data. I am investigating federations models to let users connect to other instances and work collaboratively, much like Lemmy.
So, my goals in order of priority are:
- No compromises on privacy and security
- Completely FOSS
- Read-time collaboration between users
- Asynchronous collaboration (work offline, sync when back online)
- One account to operate on all apps in this suite
- Web UI / desktop UI / mobile UI
- Minimal interface which my grandma can use, no feature-bloat
- No anti-features such as advertisements, tracking, etc…
- Self-hosting
- Federation
After Simple Notes, I plan to keep developing other simple apps, some ideas on my list:
- Simple Notes - Replacement for Google Keep
- Simple Split - Replacement for SplitWise
- Simple Chat - Replacement for WhatsApp/Slack/Meet/Teams
- Simple Docs - Replacement for Google Docs
- Simple Draw - Collaborative drawing app
- Simple Calendar - Replacement for Google Calendar
Initially I started writing my own protocol (mirror), operating quite differently from Lemmy’s, but then I realized that someone else already has developed a protocol for this purpose: Matrix. It is (optionally) E2EE, it’s FOSS, and it’s federated. So yeah, it sounds like the perfect choice to me. Also, if I pick Matrix, Simple Chat will just be a reskin of Element, so development cost is almost zero.
So, a questions for the developers on the Lemmy Selfhosted community: do you think piggybacking the Matrix protocol would be a good choice? Do you know any alternative that might be more suited for this purpose?
And a question for all Lemmy Selfhosted users: is there a simple app that you would like me to add to the list?
Bonus question: do you know of any Lemmy community where I could repost?
Thank you very much for the time you spent reading my post!
–
Link to source code on my server and GitHub mirror.
You must log in or register to comment.
Disclaimer, I am idiot with no time.
I know Joplin exists but I havent got around to setting it up.
Would you look a developing a shared notes app, Wanderlist used to be great before microsoft for Shopping lists.
I think any app for text input should have word prediction. It’s mostly for phone users but I’ve found it so handy since getting a phone where it’s well implemented (Pixel 8) I’ve quickly come to think of it as an essential component that should be available to any app.
CRDTs are good for both realtime collab and local-first.
I didn’t know this term! I knew about DAGs which are probably a way to implement CRDTs. I just read the definition on Wikipedia, it says that “[CRDTs feature] an algorithm (itself part of the data type) automatically resolv[ing] any inconsistencies that might occur [in case of conflicts]”, that sounds interesting. I was thinking to try to resolve conflicts automatically whenever possible by adapting Git merge strategies, and when impossible: either just concatenate both versions and let the users fix it manually, or giving users the option to choose line-by-line which version they want.
With this keyword I might be able to find more literature on the topic. Thank you for suggesting it!
There’s already an international open source and federated standard for Journals, Tasks and Notes. It’s called iCalendar-standard (RFC-5545) . See the VToDo component, relevant for your Google Keep alternative.
Also check out JTX Board
Pros:
- FOSS
- Federated
- List, Grid, Kanban views
- Reminders
- Alarms
- Location
- More
Cons:
- Markdown only, no WYSIWYG.
Syncing:
- WebDav server (many FOSS and selfhostable servers available). For mobile client there’s DavX.
- WebDav Encrypted sync server Etesync which is FOSS and selfhostable.
These standards and protocols are compatible with almost every software remotely related, including Thunderbird, KDE’s Kontact, and many more. And the feature set matches or even exceeds Google Keep’s capabilities.
My humble opinion is that you don’t need to reinvent the wheel, just build on top of what is already widely used, tested, documented, FOSS, privacy respecting, standards compliant, audited.
I would love a replacement for keep. I’m currently running nextcloud and the UI differences or lack of android widgets are keeping my wife from switching. Most of of your list already has great replacements that exist but for some reason a keep competitor that is current just doesn’t exist
Carnet is a nextcloud app that comes pretty close and but it needs a TON of TLC on the server side. I keep telling myself I’ll work on it but life gets in the way. If you have time and know node/php I would suggest looking there instead of making yet another note app. Nextcloud can handle sharing and has some encryption stuff as other plugins etc. Otherwise keep posting here, I would love to follow your progress.
The Nextcloud Notes app for Android does have a couple of widgets (note list and an individual note), is there widget that is missing?
For that one it’s not the widgets but the notes behavior. She wants check lists that self organize as they get checked. In the app itself she wants the post-it note like layout
Edit: multi layout that shows the contents instead of just title would be nice too
Yeah, check lists in Notes could really use some improvement for sure. Honestly, just now looking through the Github for the Android Nextcloud Notes app it looks like there’s a good deal of technical debt that has been stacking up over time from trying to bring more modern features to what started as a minimal text-only notes app.
There is a way to enable “grid view” in the app settings for the more post-it view that shows the first part of the contents, but doesn’t seem to show on notes with markdown formatting, so anything with a list doesn’t show a preview.
Sounds nice. I would use it. Keep is one of the last Google services I still use.
Hey Hey! What a fantastic idea, I sure hope you can manage!
I was immediately wondering why you wouldn’t hop on projects already out there. For example: I recently ditched google Keep // Trello for Joplin, which i think already ticks alot of your boxes, even Async collab should work (but not realtime, and you’d have to manually resolve conflicts). Maybe adding better collab in Joplin is a more efficient way to spend your time? LibreOffice also has Async collab, since I think about 6 months, so while it might be young, it will grow, and might also benefit from another person putting their time towards it.
Whatever you choose to do, I’ll be super interested to see the results.
As far as I know, Joplin is not really real-time collaborative, nor federated. I was using it a few years ago and I like it, that would be a great suggestion for people not interested in those two things.
why you wouldn’t hop on projects already out there
Ah… that’s… yeah, that would probably make a lot of sense. I think that to us developers it’s much easier to come up with our own ideas than to hop onto other projects unless they align very well with our vision. It’s quite unusual to find other people with our very same goals. I’ll be honest, I code for fun. I would like to build this tool because want it more than because I need it, as there already are both Google Keep and Joplin after all. If there was an alternative out there with most of the feature features I need, in a programming language and framework that I am familiar with, I would probably have joined or forked it. But no, unfortunately that’s not the case. My goal is not to just find/build a replacement for Google Keep, my goal is to find/build a framework that can be used to build more collaborative apps in the future. A replacement for SplitWise is my next goal immediately after Google Keep, and I assume (hopefully not too naïvely) that once I will have a starting point, making more similar apps will be virtually free in terms of time required.
Actually, I was just so incredibly lucky that the Matrix protocol seems to overlap almost exactly with all the things I had in mind. I had included a few more features in my draft for a custom protocol, but honestly adopting something already made would save many tens of hours of work, so joining something already existing sounds like a good idea. I don’t think I will be able to contribute code to it though, as it’s developed in a language I am not familiar with (yet), so if I decided to go for Matrix after all, I will just be joining as a consumer.
LibreOffice also has Async collab […] might also benefit from another person putting their time towards it
Oh, that’s actually good to know, thanks! Personally, at the moment I use Google Docs when I need to work in real-time on a document with someone else. In my opinion it’s just too convenient to just be able to share a link which can be opened with a web browser from any device and operating system. There are some good self-hostable web options for documents imho though, which are probably less resource-demanding than LibreOffice.
Unfortunately I am not interesting in contributing to LibreOffice. I have installed it on all of my devices, but I almost never use it.
Whatever you choose to do, I’ll be super interested to see the results.
I’ll take that as an encouragement! Thanks!
I think I understand your requirements more clearly now, a framework for FOSS collaborative work would be really great. I Hope you can find a way to use Matrix for it. Thanks for the explanation.
Man, I loved the google docs realtime collab functionality, they pretty much nailed it day 1 in my eyes. It’s so easy. If your project works/takes off it should be a lot more straightforward for FOSS projects to incorporate it. Hype!
Really interested on seeing this, although if I could make a suggestion, start by scouting around and see if you can adapt FOSS apps, maybe fork them and add/remove features to please your objectives and tastes.
Although I’m eager to see these through, I like projects like murena (/e/OS) that cobble together good Foss projects into a single cohesive ecosystem (without making the word ecosystem gross and vendor locked in like in most cases)
start by scouting around and see if you can adapt FOSS apps
I see lot of people are recommending me to contribute to existing projects instead of reinventing the wheel. I want this app to be as close as possible as I envision it, I couldn’t find something close enough to what I had in mind (in particular these three things do not seem to get along well: real-time collaboration, simplicity, reliability/stability). But I’ll be honest, I code for fun and I am lazy. Even if there was another project similar enough to this, I don’t think I would want to invest time to learn all the details and then more time to see if it can be adapted to my vision. Starting from scratch is easier, even though it might not be as beneficial to the FOSS community. Still, thank you for suggesting this, that sounds like it would be the smartest thing to do.
Also, I didn’t know about Murena, thanks for mentioning it!
Oh no, don’t take it as “don’t reinvent the wheel”! I meant it in the true sense that sometimes we spent so much effort and focus building something, just to post about it somewhere and getting a reply “Oh nice, it’s exactly like X project!”.
Currently I’m running NextCloud on prem, so DavX5 and JTXBoard cover most of my note taking and todo tasks, and I guess one could deploy the server-side encryption module on a NextCloud AIO on a VPS and keep everything (probably) safe and private. I’m kinda lazy too, that’s why I liked the hands-off maintenance of NC-AIO. I get notifications to update stuff, and I get regular security audits from NC itself.
BTW, never take that “doing stuff already done” is in detriment of helping FOSS projects. There are tons of examples of people randomly tinkering around and accidentally finding some huge fix for other projects. Off the top of my head, some weeb wanted to play Nier Automata at decent framerates on wine and a few years later, here we are with DXVK and all the proton stuff making most stuff playable!
Why not use Joplin? Open-source, very flexible, I run it on a bunch of devices and sync it via a EU cloud provider over S3 in an encrypted bucket…
Its not really a keep replacement. Its a good note taking app though.
Joplin is not really real-time collaborative, nor federated. Or rather, it wasn’t last time I used it a few years ago, unless things have changed now. As a developer, I like Joplin though, that would be a great suggestion.
Could you maybe work with someone together, or get inspiration from something like the docs project from the european union https://docs.numerique.gouv.fr/ ?
Moreover, I’m an “advanced user” and I want to write with markdown and/or typst. A wysiwyg typst editor would be amazing
Edit: did I mix google keep and docs? Probably. Sorry 😅
Edit2: are you looking into vjournals? https://connect.mozilla.org/t5/ideas/support-vjournal-in-thunderbird/idi-p/46295
Oh yeah, I looked into and tried some of the European collaborative docs softwares, in particular I like this Numerique’s clean Material UI interface and I’m definitely going to take inspiration from it (although I haven’t installed and tried it, I just checked screenshots). Although, if you are recommending it as an alternative to my idea, Numerique doesn’t seem to be federated, nor to be privacy-centric (well… it is if you self-host, but my idea is to let people collaborate through different servers, like Lemmy, I don’t want users to need to trust the owner of the servers they join). And I really need an alternative to Google Keep more than Google Docs. I’ll be honest, Google Keep is really well made in my opinion, I would say we don’t yet have a good self-hostable alternative.
did I mix google keep and docs
Yeah, Google Keep is more like for text snippets, links, checklists, scribbles, and images. Google Docs is a full-fledged RTF editor. I was planning to implement the two things separately. I considered implementing Markdown in the Google Keep replacement, but I am still undecided whether it might count as feature-bloat.
vjournals
Oooh, that’s nice. I know about VCard and VCal, but I didn’t know about VJournal. Thanks for the hint, this might be very useful!
IMO, the more self hostable options we have for services, the better. This sounds like a good idea to me.
Good luck with your project and thank you for taking the time to contribute to the free and open source community.
For notes, I have moved to Joplin with the option to synchronize my data using a WebDAV server. It works really well, and it has both a mobile and desktop app. If you’re interested in developing your project, maybe you can have a look at the options this provides. For example, I really like the ability to separate notes between groups, assign tags, create drawings, and the possibility to use Markdown.
Good luck with your projects! To mirror @enemenemu’s suggestion, I would also look into collaborating with the people trying to push the EU Docs alternative. Not sure if that will work, but it’s worth a shot if you’re interested :D
Oh yeah, I used Joplin in the past. Although the goals are quite different from my idea. Joplin is a great recommendation as an alternative to Google Keep, but it’s not real-time collaborative (or federated, for what it matters).
create drawings
Ah, I don’t remember this option. Maybe it wasn’t available when I tried it last time a few years ago. I guess that means Joplin is still being actively developed.
I would also look into collaborating with the people […] Not sure if that will work, but it’s worth a shot if you’re interested
Oh! I just noticed that they have quite a few things besides Docs. I mean… I probably should. My goal is for something much smaller, I just want a replacement for Google Keep, and some day in the near future, SplitWise. Joplin sounds like a better candidate than Numerique.
Thanks for pushing to collaborate, I realize that would actually make more sense than starting a new project from scratch. Not without guilt, I have to admit that I code for fun, and I have more fun working on something that works exactly as I envision it rather than just joining another project. Someday I might learn how to make some real contributions to existing software, but I don’t think this idea of making a simple replacement for Google Keep will be my trampoline to working this way. Still, thank you for trying to push me to do it!
I’ve specifically been looking for a self-hostable keep alternative for a while now. This sounds amazing, will be following the project!
same here
Nextcloud Notes or Joplin (nevermind all the other features Nextcloud provides) tick most of your boxes. They’re more productivity focused than privacy focused, it doesn’t do “zero knowledge” encryption the way you’re describing, but I don’t really understand the point of that when you’re self-hosting and the server host belongs to you anyway. The federation may leave you wanting more and the collaboration might not be “real time” enough for you either, though. If you can build something better by all means go for it.
Oh yeah, both Joplin and NextCloud are great. I tried them both in the past. NextCloud is a bit bloated in my opinion though, I was hoping to go for something simpler both to install and to use.
Most of us here on the Lemmy Selfhosted community are skilled enough to keep a computer running somewhere, expose it securely on the internet (or just LAN, if that’s good enough), and install their own services such as Joplin and NextCloud, but my goal would be to make something that you don’t need to self-host. If you can trust any server, you don’t need to host your own. I like the idea of building a network comprised of both self-hosting users, and “normal” users, like here on Lemmy.
The federation may leave you wanting more and the collaboration might not be “real time” enough for you either, though. If you can build something better by all means go for it.
I’m not sure I understand what you mean. Do you mean writing a protocol that supports federation is a very difficult thing to do? Actually my first draft didn’t include federation as that’s just too much for my skills. But I am considering adopting Matrix, and Matrix includes federation, so why not? They built real-time messaging with Matrix, so I would assume that should be real-time enough for document editing.
What I’m currently missing is a simple app for keeping contacts synced between my android phone and Linux desktop.
Excited to see what you do!
I use Nextcloud for that. On Android phone it’s DAVx5. Thunderbird can use the contact via CardDAV, DAVx5 syncs them with the Android addressbook. Fossify Contacts is nicer than the Google contacts app.
The same way it’s done with my appointments. I have also replaced the native Google calendar with the Fossify Calendar here, because it’s less annoying.
Self host radicale if you don’t want the whole nextcloud suite
Radicale is an amazing light and efficient CardDAV/CalDAV server. Pair with Dav5x on android and you are fully setup.
