I tried this Kanban plugin and ultimately ditched it. I do think it's superior to the other Obsidian plugin, CardBoard (which has some advantages too, namely that it works great with the tasks already in your Obsidian vault) but I ditched Obsidian Kanban because there was no way for me to just view one card. This is a crucial feature for me - while I'm working I do not want to look at the whole board. I want to focus on just the card I'm working on.
Assuming "kanban" is the Japanese 看板[1] ("kanban", billboard), the Department of Redundancy Department would like a word with you and Kanboard regarding "kanban board".
Incorrect assumption. This "kanban" refers to the signalling process you can use to maintain even levels of buffer inventory in processes that are too variable to do prodution levelling.
A board is just one implementation of this signalling process, but by no way implied by it.
LogSeq has suffered a lot, imo, since they took venture funding. Their focus has wavered drastically since then, shipping features nobody is asking for that I can only assume are viewed to be differentiating for other PKMs (e.g., canvas) vs making the thing performant, improving the wretched UI, fixing bugs, etc.
Obsidian may not be open source, but it reflects a sane approach to product: small, focused team, profitable company, delivering useful stuff. Clean and tight. It's got a different set of abstractions at its core (document model vs block model) but if you're indifferent to that, Obsidian is where I'd start.
> Note: Search engines will not index the contents of wikis. To have your content indexed by search engines, you can use GitHub Pages in a public repository.
Otherwise it's a lightweight alternative to Obsidian Publish. The disadvantage, that it's limited to a common subset of GFM, Gollum, and Obsidian Markdown.
You can try this plugin as well https://github.com/marcusolsson/obsidian-projects where it has a "boards" view for the projects you setup, it has a nice approach to project management without manipulating your pure notes to much other than the frontmatter data.
I tried using Linear before switching to Obsidian Kanban. I found Linear's UX and site structure to be overly complex for my personal needs. Despite following the onboarding process, I found myself clicking too many times to figure out where things are (I acknowledge that this may have been an issue on my side). As a result, I switched to Kanban because I was already using it to store documentation for my projects. For me, it meets the spec as a project management tool and is an improvement over a text-only kanban. I can definitely recommend it.
I gotta say thank you! You just introduced me to Obsidian! So far, love the app and the Kanban plugin. The plugin is basic, but gets the job done for my needs. My only gripe is Obsidian isn't open source and will probably deteriorate over the years. But I love how it's all markdown, so I should hopefully be able to transfer my data to something else when that happens.
Have you tried Joplin? It's open source. I liked it very much but ultimately ended up using a folder of markdown files and using vscode to edit them. It works great. The only issue is that long hyperlinks become unreadable.
There is! Fair warning that it's a little bit clunky. (which is to say, you may have to manually push buttons to commit/push/pull) I'm not the author, just a happy user.
I've found myself running into sync-conflicts while editing my Obsidian Vault in a syncthing folder in the past. Did syncthing work flawlessly OOB in your experience?
Yes, Obsidian by itself reads and creates files in a folder of your choice. The folder is considered a "workspace" and you can create/switch to other workspaces if you want. There's desktop apps and mobile apps. Aside from Markdown files, you can store and display images, audio files and video files (haven't tried that last one though).
However, you can choose to backup and sync your workspace across multiple devices using a cloud service. The most pain-free way to do this is by buying a subscription to Obsidian Vault. This is also a good way to support the development team, since the main product is free. When you store your workspace on Obsidian Vault, you can choose to specify an encryption key, so that theoretically nobody at Obsidian should be able to read your data.
But there's nothing stopping you from using other cloud storage services. There's even a community plugin you can find to use a Git repo for storage.
They have a paid cloud service but by default it just stores markdown files locally. Unlike many apps with that model, they're not at all pushy about their paid services in the app.
I actually ended up making a physical Kanban board at home in a central location. Relatively large corkboard using 3x5 index cards, with a few layers of backlog, in-progress lane with a hard cap of 2 cards, and about 1/3 of the space reserved for done cards. This last bit should be unnecessary in theory but I find psychologically satisfying to look at. This is for shared use by the family.
I've not bothered with a physical board, but I have a massive Trello board. For teams I strongly recommend specialized boards for projects and processes -- the process to migrate a MySQL database will be dramatically different than building an LLM.
But for personal stuff, fitting everything into one work flow is fine. If folks are wondering how to use Kanban for personal needs, I currently have 7 columns:
Dumb ideas: 62 cards
Backlog of Doom: 100 cards
Not Yet: 38 cards
Weekly recurring: 8 cards
Doing: 4 cards
Review: 4 cards
It Is Done: 1 card
Dumb ideas is a holding tank for wild ideas, usually low priority or high risk. To pick a few examples: trial adopting tiling windowing managers, buying an IPv4 block, and terraform for social media profiles are all on there. Also: "self-hosted Trello alternative." Basically things go here so I can remember I decided that idea was _not_ a good one, so I dont get into any "I thought I had a card for this" create-delete-create loop.
Backlog of Doom is your standard backlog. Lots of video games, books and software projects here.
Not Yet is basically a way of hiding things that are scheduled for later. This is almost entirely TV shows awaiting new seasons, but also a few tasks filing taxes, selling shares etc.
Weekly recurring is essentially a tighter loop of that, plus some information gathering tasks ("look for new books at the library on $topic", "review the top posts for the past week on /r/$subreddit".
Doing is as expected.
Review is a catchall for things that benefit from double checks -- financial transactions, followup communications with insurance, etc. If I wanted I could also use it for book reviews.
Done is as expected, but has a butler cleanup task to archive cards.
I've also got labels for Games, Shows, Reading, Money, and a few others.
Yea, each addition was lined with good intentions -- I don't want to prioritize the entire backlog but do want to filter out the worst ideas and the "hurry up and wait" stuff.
Let me guess: When you have a task to start say filing a pile of unfiled mail, bills etc, you start by sorting it into other piles and then run out of time, over time those piles grow, then at some point get combined then (after a suitable grieving period) you have a task to tackle this pile, so you start by sorting it into piles...
I totally understand the concern about organization paralysis, but the point is to keep it simple, focused, and relaxed. By writing things and ideas down in an organized way I don't have to track it in my head, and I can get back to what's important. Or at least procrastinate on something useful. I suppose it looks complicated from the outside, but has really only grown slowly over a decade.
Also: I have most bills on autopay, and transactions created 90d in advance in GNUCash. I do also have piles of mail in the form of a filing cabinet and labeled folders. I've actually been experimenting with letting the mail pile up in an inbox and filing it away once a month just to prove to myself the system works enough that I don't have to monitor it daily.
I also use a physical Kanban board on my kitchen fridge with Post-its. I have four columns (To Do, In Progress, Blocked, and Done) and three swimlanes (Everything Else on top, Books in the middle, and a Recurring lane at the bottom, where some cards temporarily park in Blocked until needing to happen again). The Done column works a little bit differently from a traditional Kanban, where it's reaped on a rolling basis - e.g. I work my way down the fridge, then start replacing the post-its at the top.
An interesting feature-not-bug is that my In Progress column is capped at four tickets, physically, by the ice maker.
Can you take a picture of it maybe if not too personal... I was planning to do something similar but ended up using white board with grids I drew myself.
I have too much PII on the cards, but it's just a standard cork bulletin board (about 5x3 feet) hanging in landscape orientation, and then I used long strips of paper to make vertical lanes. The middle of the board is the in-progress lane and this is marked with a neon pink paper to stand out, and a big bold note at the top "In Progress - 2 Max!" . To the left are backlog cards, to the right are finished cards. The kids enjoy moving cards from lane to lane, and we also enjoy looking at the left side together and moving cards around according to priority. Just like PMs at a prioritization meeting :-)
So far it's worked out OK but I feel it's because I am the primary user. When the kids get older or to get the SO to use it more I feel we might need to come up with private sections, or color code the cards or similar and change the in-progress rules. It's still a very fun physical activity kind of like browsing books at a brick-and-mortar store or library.
This is by the same person who made miniflux! It's a fantastic, minimal (just the right amount of functionality to be good) RSS feed reader. It's how I got here :)
Have been self-hosting this on a raspberry pi for the last 2 years, and it has been completely painless. It just keeps chugging along without needing to take a look.
Postgres makes sense if you take into account the paid hosting option (which is what I use), which requires multi-tenant support. I agree, SQLite would be way more convenient from a hosting standpoint, but minimalism on the user side does not have to mean minimalism on the hosting side.
Unrelated, but I was looking a very long time to re-discover this app (someone else posted it here some time ago). It was impossible to search for, and I haven't bookmarked it. And now I have it. Thank you!
Well, my case is that I like this app's UI (not interested in self hosted RSS), so I wanted to create similarly aesthetic website for my own use, but could not find "the picture" I had in my mind.
I recently switched from Kanboard to vikunja. Main reason is that it is more mobile friendly, which allows to jot down ideas quickly or to use it as a grocery lists.
I definitely recall seeing some other kanban implementation done in a very similar way, in that all state was kept client side in the browser, server-side was basically just a html page with some JS.
Now that I google around there seems to be plenty of such projects around, and I don't remember which one it was that I had bumped into before.
As the project is in maintenance mode (also a good sign - it has completed its objectives and doesn’t look to go crazy with new features etc) I wonder how receptive the project is to stylistic changes. Subtly rounded corners, a bit of a gradient, and a bit of padding and the cards as shown on the screenshots would look a million times better (to my very subjective eyes).
I can no longer edit, but it's worth noting that I went through my Trellinator libs and updated for use with WeKan, throwing exceptions where functionality doesn't yet exist:
I made a list of things that would need to be implemented in WeKan and/or the WeKan API to make it on-par with Trello (screenshot here of my checklist in Trello for this :)
I wouldn't say this is a detractor, based on the comment below what you quoted:
> The author of this application is not actively developing any new major features (only small fixes)
> New releases are published regularly depending on the contributions made by the community
> Pull requests for new features and bug fixes are accepted as long as the guidelines are followed
So it's more that no new features will be added by the main dev however it is still maintained. Other software like this exists (Miniflux for example) and they work well.
A Project doesn't need a constant influx of new features to be useful or even maintained.
EDIT: Turns out it's made by the miniflux dev which I respect.
In my opinion, the fact that it is feature complete and no big changes are expected is a strong point. It means if I put the time to learn this tool, it will not change on next update and get on my way.
You could technically use an IMAP server and client as a Kanban board. Make a folder per swim-lane, make a backlog, move e-mails from one folder to another as you get each item done, share the mailbox with your team. IMAP already supports searching, filters, flags, attachments. You can treat a thread of replies as comments on the original item and move them all as a group. Create a custom client to hide the fact that the backend store is basically an immutable log journal of MIME entries.
Maybe it's my age (after certain number one becomes super lazy..), but self host anything that requires me to install something is too much of an ask. Provide a binary that I can drop on my /remote/ machine and I may consider using it.
> Installing in this case means copying files to your web server.
In PHP land, once you have the web server set up with the basics, this is just such a nice easy way to deploy software. I actually run Kanboard on the same server as my WordPress blog and installed it simply by throwing the files into a directory.
The most complex thing you might need to do is add a new database somewhere but I find it so much less overhead than Docker or most other things.
It's the red waving flag of this project. I cannot understand how people are still arguing for this kind of stuff or stick with this dinosaur language at all.
I am sorry but I have such an aversion against the attitude of "we use xy because of historic reasons but hey actually it is really good because the design pattern that has proven crucial over the last 50 years imo is just overhead and you can just put some files here and there copy a bit of that and... "
> I am sorry but I have such an aversion against the attitude of "we use xy because of historic reasons but hey actually it is really good because the design pattern that has proven crucial over the last 50 years imo is just overhead and you can just put some files here and there copy a bit of that and... "
My criticism is that php was designed to be a violation to mvc. The <?php> file header already a commitment to mixing front- and backend code (HTML + php in one file) inevitably leading to spaghetti code.
The $annotation an inheritance from perl imo also a major distraction.
also Laravel, other frameworks and newer php versions might try to tackle some of the historical tech debt embedded into the language but I personally don't see a point in revival of php altogether.
I guess my point is to some degree subjectal and shaped by some major crimes I have stumbled upon when debugging. Even though the code in this repo looks fine, the mere use of php is reason enough for me to stay away from it altogether
I mean, you can do MVC in PHP if you want. I used perl long before I used PHP so the $annotation never bothered me.
I personally like mixing frontend and backend. I get why it's anathema to some, but there are simply tons of use cases where I think the simplicity it offers - being able to capture an entire page of functionality in a single file - outweighs the disadvantages because of the simplicity it offers.
For larger scale applications it's easy to make terrible mistakes leading to spaghetti code - I feel your pain about major crimes and have witnessed a few (and been responsible for some) myself. But this is possible in many languages if you're not using a framework, and modern PHP gives you way more tools to avoid it, even before you look at frameworks, which almost completely obviate the problem.
Yes, if I really want the software I will build a Docker image, but still I would prefer not to deal how to config/run Parcel/Webpack/Babel/Esbuild/Grunt/Whatever.
Personally, I think it's great that the features are there (much like you also get with GitLab), but I might opt for a more specialized solution that has been around for longer. For some folks the coupling isn't an issue and is welcomed instead, though!
WeKan is maintained, major features being added, etc.
WeKan has very polished opinionated simplified design, high focus on usability. Less duplicate menus than Trello. More shiny than Kanboard.
WeKan does not have plugins. WeKan has all features included, keeping them all working, options to enable and disable features. Many WeKan Power users use most features of WeKan.
Some Kanboard plugins are not maintained or do not work.
I've been hosting a Wekan instance for some time now and use it for work, school and personal lists. I'll have to check out Kanboard, but it would really have to knock my socks off to get me to switch from Wekan.
We use Kanboard in our institute, but its UI/UX is among the worst I've experienced. It's reliable, but extremely unintuitive, causing people to avoid using it. However, it has potential.
Oooh this looks similar to the first kanban board I used. I forget its name but I think it went out of business. I love the simplicity and the colors. Look forward to giving this a try!
Flexible and really lightweight. Been using it daily for the last year and would recommend it to anyone who is running a small-medium support team or as a planner for your own stuff.
Kanboard just works, and doesnt distract you with shiny new "features", I will definitely start using the docker and it again internally. It's just so lightweight.
Whenever I tried software-kanban tools they always fell into the abyss of things-not-currently-in-my-head and end up unused. I ended up making physical boards, either in a small part of my whiteboard at uni or by getting a small whiteboard for myself. Having it physically present and physically limited in space, just makes it so much easier to work with ... for me that is.
People in my current industry (healthcare/biotech) often use the terminology "a software" and "these softwares." It irked me at first but eventually I got used to it.
I think it's because every piece of validated software that our company uses, however small, is assigned a unique number (e.g. "SW12345"). They're not all full-blown applications, so we can't call them that. Sometimes they're tiny programs or even just scripts.
Anyway, it's possible the person that wrote this has a background in a similar industry.
I love mattermost and use it on the daily.
I got it as a self hosted version of discord but I played around with the focalboard as well.
Being a project manager I felt it held its own against many of other SaaS overpriced project boards like Monday.com.
still needs some work though but I'd recommend it.
As far as chat mattermost does look like a very good Slack-alternative near-F/OSS alternative, but these days Matrix and Zulip are also very good. Did you happen to try those options/do a bakeoff?
As far as kanban/task management goes, Gitea/GitLab/GitHub feels like the right place these days (for software companies). Are there any reasons you prefer a standalone tool over putting the PM stuff a bit closer to the code?
We use kanboard at our company and I can say it is by car the most useful yet simple project management tool. The community is also friendly. Sad that it is said to be feature complete.
Also the API is extreme nice!
I used Kanboard for a year or so and still host an instance. But I haven't stuck with it. I just could never get other people to use it collaboratively, unfortunately in 2023 it's too ugly and doesn't pass the sniff test – especially on mobile.
I'm trying out Kanboard-like tracking on Obsidian, so far it's good. Doesn't have the same level of detail, but the superior mobile story makes it worth it for me.
If you need/want self-hosting, and something more comprehensive, I recommend YouTrack (https://www.jetbrains.com/youtrack/) as an excellent replacement for self-hosted Jira&co. It's not open source, but it's free for <10 users and cheap beyond that.
I used YouTrack from around 2015 (maybe late 2015) til late 2016 at a former employer, and they only used it because a vendor of theirs used it...and everyone loved it; hence started using in in-house. It really was quite comprehensive, fluid, and lovely to work with from a user perspective. But, I'm sorry, nowadays and for some time now, unless something is open source, i either heavily hesitate or avoid using where possible. Mind you, I am not against any entity making money - in fact its quite possible to do so with open source - it is simply that i value the principles of open source. The moiment that jetbrains opens up that source code, i will give them my money to pay them to manage hosting for a YouTrack instance. Separate of my ethics/values, again, i can not deny that the product was rocksolid in so many ways! ;-)
I understand the reasoning, and YouTrack will fail the open-source test. I mention it as an option because we needed a self-hosted project management platform (for regulatory reasons), and I could not find any open-source options that I could put in front of non-tech people with a straight face. FOSS surprisingly skips over this area (maybe because many engineers seem to dislike "project management").
You're not wrong, sadly! While i think there are FOSS options for managing projects, i feel they work only effectively for individuals or very small teams. Maybe also if the small team works in a small (or non-profit) org where there might be less need for corporate style work constraints. But, once you go up a level, and the team gets bigger, or more sophisticated (tech or non-tech, doesn't matter), or even if a small team but operating in a bigger org...that is when many FOSS project management stacks fail to live up to hype. I say as a staunch FOSS advocate, sadly!
What I really want is a way to get a Pivotal Tracker-looking board in JIRA. Jira for the extra features. Things I love about tracker:
- vertical view of the backlog
- automatic timing estimates based on velocity
- add deadlines and get automatic feedback on how you're going towards it
For some reason I can't get this same clear board with a pull flow in jira.
PT is so much better than Jira, and I doubt you'll ever be able to replicate it in Jira. But once a company grows to a certain size the forced migration to Jira is inevitable. At that size the pointy hairs hire people who's only job is to twiddle bits on Jira, and they wouldn't have anything to do if you didn't migrate to Jira. Your only real choice at that point is to decline all meetings where jira is involved.
Hey, you might want to check out Rootspace (https://github.com/clearview/rootspace). We just recently released it as open-source, although it was an internal project for a long time.
Tangential: What are popular options for self-hosting project management tools these days?
There used to be Trac or Redmine but people seemed to move on and have settle on hosting solutions (which makes sense). But there must be some self-hosting alternatives. Maybe GitLab? Any tips are welcome.
I used trac in its most basic way (just tickets with comments) for the first time in a long time for a project last year, because a vendor asked to use it. It was really nice especially when it was required to look back a discussion or decision. I don't mind any of these tools, even JIRA if it's not been setup to micromanage, but even then it's not been my experience that you can look back at something from last year and get good information. Yes, it's immutable but as people come and go it's very hard to have the configuration of projects, workflows, etc be consistent enough over time to trust you're looking at the primary source of information.
Self-hosted Gitlab is pretty nice and is a (the?) current de facto standard. I've not used any of the CI features but the wiki, git handling, and issue handling/kanban boards feel like a nice sweet spot of functionality vs simplicity. I have also not used the features related to epics/gantt charts, etc but I personally think a good project manager makes these separately instead of trying to automate reporting based on tickets.
In the spirit of sharing, I wanted to mention a project I'm working on, called JustDo.com . We've been developing this enterprise-grade project management and planning tool and have seen validated product-market fit with some Fortune 500 companies.
We believe strongly in the power of open source and we're preparing to go 'Open Core' by this summer (See Community Edition in the pricing page to see what will be Open Sourced). Our intention is to give back to the community and enable more people to benefit from our tool.
If anyone is interested in exploring more about JustDo or participating in our open-source journey, I'd be more than happy to chat or answer any questions you might have :)
Thanks for sharing. I’ve been lashing a bit with openproject and tuleap, and while the latter is fascinating the lightbulb of what kanban provides me over and above agile has me craving it.
Used it in a project, worked ok. One problem back then was it was desktop use only, using it on a tablet or phone didn’t work well, perhaps this has improved since then.
It's not perfect on a phone, but I use it on my Galaxy Android phone occasionally. AFAIK it works better in desktop mode, depending on what you want to do.
Yes, you can make swimlanes in Kanboard. I've used it for several years now (self hosted), and can recommend it. Migrated from Trello some years ago. Mostly using it for small personal projects, but it seems to also scale well and has a lot of functionality.
It doesn't meet today's design trends but it works perfectly for tracking and updating issues on projects. when I was a consultant I used it to track time and export that to make invoices.
I like pretty things, but for Kanban the tool’s appearance is perfect IMO. If anything, I’d want it to be duller: I want to spend as little time looking at my task management as possible, and making it prettier only serves that purpose if it’s more usable as a result. But this is already perfectly usable.
I think the landing page could use a little typography and spacing polish, but again it’s perfectly usable. It’s worlds better on my phone than ~99% of what I encounter where the lede includes “free and open source” or the target audience ≈ most of HN.
The only thing I really could criticize (with the tool, as presented) is that it’s susceptible to the same problems every software kanban implementation eventually succumbs to: every Trello eventually becomes a Jira. IMO it’s already too featureful (and I know some other commenters disagree here), not too much to rule it out for me but the real risk is there’s a very clear path for feature creep that would degrade its appeal (to me).
Please no. The trending way to make this "Pretty" is to round all the corners, add tons of whitespace, remove text labels from buttons for bespoke iconography, hide half the features in a hamburger menu, and add a full height side gutter for 3 icons.
No thanks, give me a tool that shows me content and options I can apply to it and forego the rest.
do you want a helm chart for this? the "requirements" page scares me, although I suspect this whole thing could be launched with a docker-compose / helm deploy very easily.
I ignored the requirements page entirely, downloaded it as a zip file into my existing (PHP-based) web server, unzipped it, and it basically worked immediately.
I think this would be the case for anyone else running a relatively modern existing PHP setup, and for everyone else they can just use the Docker image.
I wish people would appreciate other peoples effort a little bit more. As far as I can tell it is a one man project that has been around for a while. It takes a lot of effort to do this.
It is a valid concern though. It's very common to see technically genius projects, that don't get enough attention because their project landing page looks horrible.
I kinda don't understand how people can be so skilled, and put so much effort to make a nice thing, and then completely skimp on the design.
E.g. here it wouldn't even take much. It's mostly just padding issues, and having sharp and rounded corners mixed, and maybe the color palette of the tasks could be improved.
I actually like it, apart from maybe text needing a little bit more padding when close to the edges of the cards etc.
The UI is minimalist without feeling like it's lacking, it's usable and the performance is great because it doesn't try to have lots of fancy gradients or animations for the sake of it.
Still using OpenProject because of the nice multi-project feature set there and Jira having rotten my brain, but Kanboard is easily 10x times faster than the sluggish OpenProject experience.
Been a while since I used it, but as I recall it was very simple to customize the design. My opinion is that, aside from fonts, functionality trumps design, and there's typically a tradeoff between the two.
https://github.com/mgmeyers/obsidian-kanban