Having frequented several cafes in London, I was struck one day coming across a particular low back chair that was more comfortable than my work provided Aeron chair! I went searching on Amazon and found the following to mimick it https://www.amazon.co.uk/gp/product/B089QYLY69/. Never looked back.

Except I lower mine even more than in the image above...

the Capisco is the perfect chair for the https://media2.giphy.com/media/v1.Y2lkPTc5MGI3NjExaHNpcDNic2N1bGFmazU5N3pxcnI2bmVjYWhpbGw1NTRicmJqN3VpeCZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw/PJWKBfhzrwKtO/giphy.webp

I got a Capisco last month It's not really a chair but a stool, the back rest is a transitional aid (since not using it needs developing some back strength) Its overall paradigm is 'active sitting'. Seems a good idea, personally I'm parking it for now though as I already do lots of sport to exhaustion, so I cannot randomly add yet another activity

Next purchase might be an elliptical machine (or vanilla treadmill, IDK), it's also active but more holistically and without letting muscles shut down or under pressure

yes the back is often a pain, but I what I like is how natural I change postures when having long-ish sessions

btw, the Capisco experiment prompted me to try a back cushion. the Tempur one (pictured) really improved my sitting in either chair. They're an unparalleled brand, the quality of the material makes all the difference between an idea that works and one that doesn't Now I'm at my best sitting experience in years with the back rest, donut pillow, and small reversed Bosu ball for the feet (promotes some gentle movement)

I trimmed the Capisco arm rests btw as I like my arms active/free 🙋

I have the Capisco chair in the office, and love it. I'm going to buy one for my home office soon. 🙂

wow you literally cut it @U45T93RA6

Wow, I'm way behind in the sitting game

https://github.com/org-roam/org-roam in Emacs.

big Logseq fan (written in clojure!) https://logseq.com/ considered trying to work org-mode into the mix as im getting more into emacs, but logseqs queries and block-based organization is just really great for me

What about for an entire organization?

What we use at work: Confluence. What I'd prefer: anything but Confluence 🙂

If you're interested in Roam, Logseq, and the like, I've also heard good things about Obsidian, where you can work with teams: https://obsidian.md/

Iiuc Roam also works for organizations @U04V70XH6 I have colleagues who say even Confluence is preferable to Notion. Send help 😄

Oh boy... Re: Obsidian -- I like that a lot for individual notes (I switched to it from Logseq after several months). Hadn't looked at it in a team context tho'...

I have much curiousity about the Notion fatigue. We're using that, since before it was cool too. Notion is the best approximation of a whiteboard-that-stares at you. Then it's our job to wrangle that whiteboard into what we want. And that's hard, and messy, and a continuous struggle. Which is a Good Thing I think.

Sharing knowledge, discoverability, connectivity, are all terrible in notion. Tables and hierarchies to not map to thought

I personally much prefer Logseq but that does not scale to an org

Obsidian it's the emacs of notetaking world. You spend more time setting up and beatifying than taking notes at all 😄

Jokes aside, Logseq is amazing, is what made me interested in clojure, but right now I'm taking a look into org-roam. Issue with org-roam is mobile syncing I think. About collaborative, I'm following Anytype. If you don't need calculations and so on could do the job

Not an off the shelf product, but actually works for a small team: Word documents in a shared folder, wired up to Hiccup templates that use the text from sections of the word documents to generate a static website. Word documents can be redlined, have a decent workflow for collaborative editing, and don’t have any learning curve for non-technical users, unlike basically everything else that might fill this niche. This is the only approach I’ve found that both “scales down” to prose-heavy docs written by a non-technical author and “scales up” to complex technical writing that requires pulling in data from external sources and visualizing or generating structured documentation from it.

you will have to figure out how to put it behind a login wall, but lots of static website hosting providers support oauth workflows now

For personal usage, I like org-roam (for things I intend to put on the web) and Roam (for stuff that's "too small to track in Git"). I used Tana (https://tana.inc/) a big in a team context this spring - want to use it more. I think that's more of a Confluence killer than Roam - specifically because of its support for metadata and to build table views on that metadata.

Might be more specific: http://backstage.io

I have an irrational hatred for Microsoft Office, but nothing has beaten Sharepoint for me. Unlike Notion you can have teams, an actual heirarchy of folders, fine-grained permissions, and host custom sites with no code (so accounting and management folk can create their own dashboards and metrics pages). I have no idea why so many startups use Notion. It's nearly impossible to organize and search, and for some reason updates / history require a more expensive plan than the one we currently pay for. But I simply have to live with Notion. tl;dr: sharepoint is awesome. if your organization already uses office then strongly consider sharepoint

➕ you can repurpose SharePoint to be a generic static web host if you rename your .html files to .aspx files and put them in a folder that isn’t a “document library,” which gets you authentication and access control for free

What's wrong with good old wikis?

Wikis are where knowledge goes to die, in my experience 🙂

my (large) company is in the process of switching from Confluence to Notion, which is a step in the right direction IMO. We also still use good ol’ Google docs a ton for design docs, run books, etc. When you’re dealing with a large org, you kinda gotta go for a low common denominator tool since people’s preferences around this stuff are so wildly different. The value of standardization is probably greater than finding the best tool for the job

My company contracts with american fantasy author Patrick Rothfuss to turn all our notes, todos, and documentation into a cohesive, entertaining stories. We keep him fairly busy, luckily, he has no other pressing projects.

Is he going to take seven years to finish a single story?

I found wikis is where information dies the least, because it's more discoverable. It tends to have good search and no restrictions and things are a bit better organized. But I've never tried Notion.

Google docs, word, note taking apps, all that I find it's very hard to look for information on those, it tends to die with the person who created the note or document

google spreadsheet confluence was not the worst but some sort of private wiki would be cool preferably built on clojure and with clojure syntax support in the case of code management

Wiki built on top of #C035GRLJEP8 🙃

We do on-prem confluence, imho it's alright. Except for search and the inherent page hierarchy, which may be one contributing factor that we have a lot of duplication. An obsidian-like graph view would be useful.

This is why I like Notion: Blocks are addressable and freely moveable. Pages are blocks. What just happened: A private thought written down in a private Notion page, was discussed together and then moved into a public space. And all references to that private thought remains intact. Notion allows us to refactor our information and shape to our process. I value that.

sounds like many people here may be interested in #C05N8TX4YPM!

i do have it starred on GH, still waiting for the source to open up :^)

Seconding roam (org-roam) or obsidian. The network map is a much more intuitive way to find documentation. I hate hoping a text search strikes gold, or staring at TOCs looking for something relevant.

I wonder if Notion has considered using their rushed forced AI features for semantic search of documents. That would solve most of the magical text search problem. (I say "most" because word2vec with a single language corpus won't help in multi-lingual knowledge bases or knowledge bases with tons of custom terminology)

Honestly, in order to really help here, I would have to know what information your trying to keep track of in the first place.

We also use Confluence, and I’ve found it has the same problems as any other tool I’ve tried: it does nothing to prevent the kind of thoughtlessness that stops it from being useful.

In my limited experience, someone must be in charge of the knowledge base. it's gonna be in the job description and active duties. otherwise, the knowledge rots and people lose focus or stop caring.

I'm confused how any one person could be in charge of collective knowledge. Though i guess, someone, could be everyone.

If it isn't some specific person's responsibility to keep it up to date and organized, then it is no one's responsibility... And it won't be kept up to date and it will become a mess...

I think it's more about responsibility for the overall structure than all the Individual content. Making sure things are in the right place.

Right, it's not on the Knowledge Base Owner to write everything necessarily, but to keep the pages organized and correctly interlinked, to create templates, to clean up messy pages, and to apply top-down pressure on teams/devs to write documentation. At my last job, I'd get assigned a body of work with a half-written stream-of-consciousness "project plan" and spend hours and hours bugging the authors to clarify stuff as I stumbled through it. Having someone to say to the project manager or engineering manager "Hey, this isn't quality work, please revise." would have done wonders for me and my team mates

When I joined, the "how to set up a dev environment" page was split across 3 pages, each had contradictory info, lots of notion comments in the side that had arguments and discussions, no clear concensus of how to start from scratch lol. I spent a bunch of time over my first year cleaning it up only to realize that I was in the last wave of hires before they had back to back years of layoffs but at least the "New Hire Documentation" looked great

That is essentially where I’m at: I am the Confluence janitor for our team. But more to the point of the original question, at this point I’m 100% certain that teams will rapidly get fatigue with any tool. Information Architecture is hard, never-ending work. I have my own opinions about what a given tool should do, but I also know from first-hand experience that will only work for a handful of people and everyone else will quickly grow tired of my solution.

It has never occured to me that knowledge bases could have a designated owner/janitor responsible for them.

> When I joined, the "how to set up a dev environment" page was split across 3 pages Ours is: install git, add your SSH key to BitBucket, download this script from BitBucket and run it 🙂 We're always looking to automate more pieces of setting up dev (and new servers) But we do have slightly more information in Confluence than that. And we have a page about how to do various deployments (there are a couple of scenarios where deployment is still scripted rather than fully automated). And we have a page describing what's running on all our servers, including their (automated) patching schedule. Mostly, we try to keep dev stuff in the repo -- but for a new hire, they need to know how to get the repo setup on their machine. But I mostly just hate wikis and always have. We have a Confluence "space" for each department at work. Some teams love wikis and have copious, well-organized docs. Some teams treat it as a giant brain dump you have to constantly search for anything. Our team is really kind of minimal.

I tend to enforce to have one big giant page. We just call it the runbook. And I tell people to put everything on this one wiki page. Everything that is meant for devs on the team goes in one page. No categories, no sub-page, no links to other place with more info. It's great. All the info in one place, you can just Ctrl+f to find what you need. You can copy/paste it into GenAi and ask it to help you with things. Everyone starts by pushing back on me, and then once they try it, they're like. Ok this is great.

That's the "Big Readme" style, which I think is pretty great

Actually, we did made it a tad better. When we create a package, in the actual README.md, we just put a link to the wiki page for it. And then in our "runbook" page, we can inline other pages. So we inline the wiki pages of all our packages. So it kind of goes, in one big page called Team X runbook:

# Setting Up

# OnCall

# Sample requests

# Packages

## Package 1 (and those have their own page as well, which the code links too)

## Package 2

# Tips

# Misc

# Links to Past Designs

And everyone maintains this one page, and follows the pattern of one wiki per package which is inlined in the runbook.

This helps when we transfer package ownership to other teams, then you can just remove the inline for that package.

We have a Landing Page, which is for like business people/stakeholders to know what the team does, how to reach out for requests and all that. And we have a Onboarding page for other team's who need to use our stuff. That page is similar in that it's a big page with everything any outside team might want to know about everything the team allows them to use.

So on the wiki it just goes: http://company.wiki.com/our-team/landing http://company.wiki.com/our-team/onboarding http://company.wiki.com/our-team/runbook http://company.wiki.com/packages/package1 http://company.wiki.com/packages/package2 http://company.wiki.com/packages/...

And there's a way in our wiki to have like a Menu header. So at the top you see: Landing | Onboarding | RunBook