Roam: Why I Love It and How I Use It | Hacker News Reader

Roam: Why I Love It and How I Use It

nateliason.com

288 pointsyarapavan169 comments

I like the author’s enthusiasm for organizing personal information. That was me from about 15 years ago to about 3 years ago.

I stopped spending time and effort when I realized how little I used my organized notes. Specifically, I used Evernote and had my personal information wonderfully organized, but rarely used it. Now I use something simpler:

For work: about 20 Emacs org-mode files for various topics. I use grep to find stuff.

For personal: at different times I have used Apple Notes or Google Keep to maintain todo lists, shopping lists, notes on present and future writing projects, planning travel, etc. Both support search. I spend very little time writing these notes and they are useful. A few times a year, I spend a few minutes deleting notes I don’t need anymore.

Anyway, thinking about cost/benefits, not spending much time organizing my own information makes sense to me.

The turning point was spending months of part time hacking writing my own Evernote’s clone in Clojure in the backend, a Firefox plugin, and a web app. It was a fun project, but mire fun writing it than using it.

I use my notes for big ideas that serve as fodder. So, I don't really keep notes on like "Here is how to do X in Javascript well" but instead "here are a collection of clippings on interesting LED art pieces and a link to my parts inventory".

the pitch for Roam imo is that the #1 challenge of modern business is scaling communications. All large companies have dysfunctional communication and if we can get a 10x improvement here that would translate to getting the same amount of work done with 10x fewer people which is a whole layer of middle management just gone. Or alternatively you can continue to barely keep it together at 10x greater scale.

> the pitch for Roam imo is that the #1 challenge of modern business is scaling communications

I completely agree with this, but I think it's naive to think this is fixable by some tool.

Scaling communication is just innately hard problem - the most efficient you can get with communication is if you can keep everything inside one brain, next best thing if you have one colleague with whom you talk all day, then it's still quite OK in a small team, then you get multiple small teams, then multiple departments, located in multiple locations, continents etc.

There can be tools and processes to make this a bit more efficient but you won't remove this problem short of doing some sort of telepathic brain interconnect.

And existing tools (like confluence etc) are already getting getting quite close to the local optimum, of course you can improve upon them, but nowhere near the claimed 10x improvement ... that's just bullshit marketing talk.

Confluence is light years away from the potential of this kind of tool I think, well over 10x. And there's a superlinear payoff with really good tools, because of network effects both in amount of users and the quality/quantity/timeliness of information that good tools attract.

I work with large knowledge bases daily and the choke points just aren't in the tools - it's the humans. People generally suck at writing - they often have very incomplete/skewed information, write only from their own perspective and are not that heavily invested in having great knowledge base (as opposed to solving their own immediate problem).

True domain experts who understand things typically don't have the time (and/or skill) to write good documents.

Documents don't get updated or in better case there's a series of "patch" documents describing the changes but not the complete future state.

I don't see how these major issues can be fixed by technology.

This could be largely mitigated by organizational changes - hiring dedicated technical writers / "knowledge managers", but that's just a cost center for companies so it essentially never happens.

I suspect there are a lot of different user bases and we're not thinking about the same one.

There are existing organisations and institutions where people want to find tools like these and are excited about the prospect, for example expert organisations where cooperation and single shraed information base in a project is very valuable and the users are good at writing & collaboration, used to real-time tools with fine grained tight knit collaboration like Google Docs / Slack.

Techology can grease the wheels and provide multipliers for motivated users, sometimes very large multipliers, but of course not every person is suited for it.

> Anyway, thinking about cost/benefits, not spending much time organizing my own information makes sense to me.

I had an employee that worked for me that used to save everything! I came from an airframe manufacture in my past that automatically deleted emails more than 30 days old and twice a year had a cleaning day to destroy any prints or notes in/around our desk that hadn't been used in the last 6 months (this was all for lawsuit reasons to reduce discovery). So I had him explain why he kept everything and spent so much time organizing it. His primary reason was that if a customer every came back and claimed they wanted something different from the original request that we could prove them wrong.

I simply asked him, "and if you can prove them wrong will that change their mind, or just piss them off that you went out of your way to prove them wrong?" He tried to argue about costs and requirements, and I reminded him that if anything was a contractual obligation then it needed to be captured in a requirements document and a change request put in to make a change; if on the other hand it was more of an informal request for support or information, if they change their mind just roll with it no need to prove they already asked for it.

Roam appears to remove the time and effort needed to organize notes. Just write. Links will happen. No need for silos, no need for hierarchies.

They are removing the friction and effort. This has huge implications and potential.

I am with you on not going and looking for notes (I don't often browse through my photo library either, which is a close analog).

This really feels like it's onto something special if you can let go of the evernote paradigm.

Yah, but Eastgate's Tinderbox has been doing this for damned-near 20 years, no? https://www.eastgate.com/Tinderbox/updates/Tinderbox8.html

This is exactly me with Evernote, and my same thoughts reading this.

I wish Evernote had some sort of stats on notes: I'm sure I would find that 90% of my notes have never been opened. Some of them (e.g. warranty numbers, etc) are there just in case they're needed, but the majority will never be opened. I probably use the same 2% of my notes (< 30 notes) 90% of the time. (Lots of invented stats here, yes.)

The article describes an incredibly easy content-creation system, but one where you can easy create much more content than is useful. (It reminds me of the ease of taking hundreds of photos a day with smart phones, particularly if you're a new parent.)

Even when you want to find the information, I think this lack of structure makes it hard to do so. For example, the author has a note that looks like this:

    * Had a really good [[Super Tuscan]] [[Red Wine]]  
      * [[Dogajolo Tuscano]] from [[Tuscany]]

My understanding (from looking at the other examples) is that if he went to the [[Red Wine]] page (presumably to look up a nice wine to buy), he'd see

    December 3rd, 2019
      * Had a really good [[Super Tuscan]] [[Red Wine]]
    Some other date
      * Bought a [[Red Wine]] for [[so-and-so]]:
    Yet another random note about other stuff:
      * Buy some [[Red Wine]] for [[whats-his-name]]

Even the links that point to useful information don't have the information right there -- you have to click through for each link. And some of the references may not have anything to do with a good wine recommendation at all. So you have to edit the [[Red Wine]] page to make it a useful list, but that's what we do with Evernote already, and the linking part wasn't that useful.

Not necessarily, you can filter the backlinks to [[Red Wine]] from within the Red Wine page to only include other tags, like Super Tuscan.

You can shift-click them to open them in sidebar (where you can edit them the same way as in main view). You can open unlimited number of pages/blocks in the sidebar. And if you find that they're not useful for you anymore then you can simply remove or unlink them. Without ever leaving Red Wine page.

It's funny how almost everybody here tries to compare it to other things. Yes, you can do everything even on paper! Zettelkasten was done first on paper, GTD was done first on paper. But this is new kind of software, which promotes new workflows and ideas about using your thoughts and data, and if you want to compare it to existing solutions then you should compare it to TheBrain or ConnectedText. It's not your typical markdown editor / outliner with some new features - you can start using it like that, but you will change your workflow after few hours or days in ways that you can't simply imagine right now :)

I hope we will have more software with such flexibility in the future, competition is always good, I might even reimplement this myself as my private hobby project, because it's so useful and I can't imagine going back to normal note editors if something happens to Roam dev.

Perhaps this is all true. My main point, which was lost in my example of [[Red Wine]], was more about it being too easy to create new pages.

In Evernote, I've tried to get away from creating notes on every random new thought, and started towards maintaining more useful large notes -- a list of books I want to read, instead of 50 pages that mention a book all tagged with "#books-to-read", for example.

To me, this seems like moving back the other way -- it encourages you to create those 50 pages, because they all can be "linked" to each other.

I guess I'd have to try it to know, but I feel like, for me personally, information would be lost.

It will not be lost. Don't only give it a try, but force it to work for you and it will. I changed my process of writing in Roam multiple times, each time it was more optimised for me and weirder to explain for other people. I have GTD system that finally, after years of trying in every other existing software, works exactly like I want. It looks way different than any example of "GTD in Roam" I found so far - those examples could be reimplemented in Dynalist/Workflowy/other outliners - mine can't, it requires these mentions on the bottom and editable block references from other pages in my main dashboard page. But I didn't design it, it created itself. Explore and you will find your own ways. If you like simple lists then create simple lists without any links, that's perfectly fine, you can always change things later. I'm personally doing way less linking than author of that blog post, I don't feel like I'm forced.

You can do there much more than that blog post describe and Roam docs also sucks in explaining what else than writing you can do so it might be a good idea to look for videos in Youtube, Roam dev created some demos and was also interviewed with demo presentations.

One of the devs here - totally true, our help docs and landing page don't really show half of what you can do. Help docs have a lot if you dig around though.

Heres a link

https://roamresearch.com/#/v8/help/page/1wnq-ZAAN

Really appreciate users like pps who do the work of figuring out how to make Roam work for them at this early stage.

(1) How close are you guys to a mobile app?

(2) What's the monetization strategy?

Thanks!

That data feels closer to prolog, than to text articles. Since search was the biggest problem with evernote, i'm curious what machine learning could do with the data in roam.

Datalog :)

For me the notes app is almost like a workspace - I need to figure something out, I write it down, start to break it down, describe, analyse - this process helps me to understand the problem. The fact that the artifacts of this process will not disappear is a nice bonus. A lot of the time I don't ever read them again, but I got the main benefit (understanding) already anyway.

It's a bit similar to writing a journal - just the act of writing it down has big therapeutic and introspective effects.

For me, the act of organizing the information actually does not take too much time but that might be caused by the fact that I typically don't obsess over correct categorization or do this "refactoring" when I'm bored. Main tool to find notes of interest is search anyway, categorization for me serves as a way to find related notes.

> The turning point was spending months of part time hacking writing my own Evernote’s clone in Clojure in the backend, a Firefox plugin, and a web app. It was a fun project, but mire fun writing it than using it.

Lol, sounds like me, except it's been almost 3 years since I started (https://github.com/zadam/trilium if anybody's interested)

Just as FYI - Roam is built in Clojure/Clojurescript and we're hiring

Is there a link to your hiring page? I'm not seeing one on roamresearch.com

No hiring page yet, just dm @conaw on twitter for now

I've got almost 30 years of notes ... those notes have proven prior art for patents I didn't believe in and establish the date of invention for work I've created. There are good reasons for scientists and engineers to keep a notebook but the traditional paper version isn't nearly so searchable as something like the author is describing.

P.S. Sometimes it's funny to see a grocery list from 15 years ago and see how my tastes have/haven't changed.

Yep, can resonate a lot with that myself. I fell in love with GTD 20y ago and spent years trying to work out the best solution for _organizing_ thought, rather than putting all that energy into actually acting on thought. It was my hoarding & librarian lizard nerd brain gone into overdrive.

It’s similar I think to what I did back in those same days when I switched briefly from Mac to PC (from the pain of MacOS 9 during the super early days of OS X), and spent so much time building my own PCs & installing configuring tearing my hear out re: CD-drivers et al plus just getting to a workable state over and over with Gentoo Linux or whatever flavor du jour — I did stuff FOR the computer instead of actually doing valuable stuff WITH/ON it.

Nowadays I use mostly Apple Notes for pretty much everything (even though the backup story still sucks). But I’ve always been a huge fan of ”outliners”, it seems that it’s the closest to how my mind works. Still use OmniFocus for some personal task management (and OmniOutliner before that), and have used Dave Winer’s very nice little outliner web app off and on for years, just to jot down complex thoughts quickly.

But Roam is definitely the thing I’m most excited about currently in this space. I find Notion.so and the others much too complex for most my needs (even though the inline database is a wicked cool idea!).

And the fact that Roam is a small ClojureScript shop as well is just... PERFECT :)

I’m intrigued by Roam, but I can’t see myself using it as long as it’s browser & cloud only. Also, good typography and visual design is critical in any kind of note taking app, but in Roam that barely feels like an afterthought.

Something like this but as a plain text based, native Mac/iOS app with stellar typography... that would be a dream.

Ditto. I was actually thinking of trying Roam until I read there was no mobile app. I like the idea of a “second brain”, but for me that has to be something that never gets detached from my first brain. It has to be accessible at all times, for both reading and writing. I currently make do with Apple Notes, but I’m interested in finding something more organized…

A second brain should also be as private as my first brain – ideally end-to-end encrypted, which I guess Roam is not if it’s a webapp. Too bad. (Apple Notes isn’t end-to-end encrypted either, but that’s another reason I’m looking for an upgrade.)

Even though it's a webapp it can be end-to-end encrypted. Fortunately, they are planning to do it, but I am not 100% convinced that all the function will work on e2e notes.

Just in case the NSA is keeping special tabs on your favorite recipes and approaches to meditation.

The NSA, or, y'know, the app's creators.

I've been using Inkdrop, mostly because I like my notes E2EE. It's worth a look: https://docs.inkdrop.app/security

The product is improving quickly - I'm happy it doesn't have the weight of two native mobile apps to drag.

> Also, good typography and visual design is critical in any kind of note taking app

The org-mode fanatics may disagree with this

Good typography comes in many shapes. If you use org-mode you can tweak it to your heart’s content.

Same with web app. I'm using Stylus for my own tweaks + Dark Reader.

Spacemacs org layer is a good place to start.

Yeah those are definitely the big limitations right now. From talking to the founder the desktop app is on its way, mobile will probably come later.

I've been using Drafts as a temporary holding spot for things to add to Roam when I'm on the go.

Dynalist offers free account and works offline on all platforms, and was the inspiration for Roam, one can also try that for that temporary holding spot.

Seconded and I'm glad I saw this before I got too deep into it. I'm wary to try even something totally open source because of course their is the adoption cost (do you move all of your old notes over? do you have a split brain? Too much waste time in tiddlywiki for me is a harsh lesson).

That said, I still appreciate the effort people spend in thinking about these things, but a cloud-only offering (cloud-only and 3rd-party hosted, I should emphasize) is sadly not for me.

From privacy policy - "As a condition of your use of the Service, you grant Roam a nonexclusive, perpetual, irrevocable, royalty-free, worldwide, transferable, sub-licenseable license to access, use, host, cache, store, reproduce, transmit, display, publish, distribute, modify and adapt and create derivative works (either alone or as part of a collective work) from your public Content."

So "public content" on Roam is considered essentially "public domain content"?

It's not public domain; only Roam gets the permission.

They demand an irrevocable worldwide license, with the ability to transfer the license and to sublicense according to to the text in the post you're replying to.

That's still not public domain, though. I doubt Roam will sublicense to me (for example).

That is standard legal boilerplate that allows them to host/display content that you mark as public. All sites that share user generated content publicly have similar clauses.

That doesn't explain modify, adapt, or create derivative works.

It does, though. They need to be able to update the saved data to be compatible with updated code.

It's a common problem with legalese. Even straightforward things sound scary without a lot of additional text explaining everything.

I think "adaptation and derivative works" could apply even for simple format conversion like Markdown (what you authored) to HTML (what you see) or even image thumbnails of pages (?)

Medium's policy doesn't seem to ask for rights to create derivative works and such. I found the verbage unnecessarily broad for what i know is likely to be narrowly applied. If RoamResearch decides to declare that all public content on Roam is relicensed under public domain tomorrow, that would seem within the scope of the policy as stated, though perhaps not within the scope of applicable law.

edit: I see that notion.so also has the same wording.

Lots of Roam features, like the visual overview, are derivative works. I don't think Medium has that.

Maybe I am from a different time but I simply cannot imagine entrusting my private thoughts and knowledge to a cloud service ( aka someone else’s computer) without end to end encryption.

They've said E2EE is coming out pretty soon, it's a high priority project

I personally think it depends—there are multiple levels of knowledge that you can store in a notes app. I personally would love something that could store my lecture notes and provide me the freedom to take them any way I want to.

I wouldn't care if a company had access to those things. But yeah, definitely not personal data.

For Windows users who prefer native apps and control over their data, Connected Text (paid, closed source ultimate kitchen sink with 30 days trial)[1] and WikiPad (free, open-source wiki engine)[2] are good alternatives to Roam.

Personally, I use TreeSheets [3] for note-taking and try to organize them as nodes in yEd [4] when I want to gain insight, but that quickly grows out of control and becomes unmanageable. Auto-linking and graph overview are extremely powerful features in this regard, but my gripe with e.g. Roam's approach (judging by screenshots) is that you can't visually filter out all the nodes at ones and "walk" the paths in a filtered graph -- that's why I consider switching from yEd to VUE [5], which has more decent filtering/search/layering/grouping options.

In general, [6] and [7] have good coverage of existing note-taking and mind-mapping tools.

Does Roam have any data export options though? If so, I might give it a try.

[1]: http://www.connectedtext.com/

[2]: http://wikidpad.sourceforge.net/

[3]: http://strlen.com/treesheets/

[4]: https://www.yworks.com/products/yed

[5]: http://vue.tufts.edu/index.cfm

[6]: https://pauljmiller.wordpress.com/category/software/note-tak...

[7]: https://pauljmiller.wordpress.com/category/software/mind-map...

Co-founder of Roam here. Also huge fan of treesheets, do recommend.

We export to markdown and JSON.

Cheers!

This was very interesting for me to read. I am a heavy user of Notion and recently documented my own knowledge management practices [1].

I understand some of the Notion criticism, but I think you can get around most of it. Instead of relying on the tree structure, I organize pretty much everything in databases, which can have relations to each other. This way, you can have bidirectional linking and the user experience for creating these links is very good.

[1] https://tkainrad.dev/posts/managing-my-personal-knowledge-ba...

No, it is completely different experience to do that in Notion and (have it automated) in Roam. I was heavy Notion and Dynalist user, now I'm completely sold to Roam. Try it for a month with some hobby project and to log your daily life, it's crazy good, after few weeks of use I'm still in awe and I find new, better workflows every week.

I will leave the argument which one is better for the marketing people of the respective companies. My post covers also other tools that I use, such as GitLab and Workona.

Just wanted to point out that Notion is very versatile and that there are ways to achieve bidirectional linking.

> I was heavy Notion and Dynalist user, now I'm completely sold to Roam.

What scares me about this comment is that, in there lies the possibility of a better notes app coming along, and having to deal with porting the data from one place to another.

I've jumped around between a whole bunch of different note-taking apps and methods (Markdown notes in vim or ST3, Evernote, Keep, gdocs, Nuclino, etc.) and have mostly settled on Bullet Journalling. The act of writing itself is helpful for memorization, threading and indexes helps with information location, and the mobile app lets you take pictures of your notes and link them to others (across notebooks, etc.).

I am tempted to try Roam regardless because of tagging, but the lack of some way to sync with my more important physical notes kind of turns me off of the effort, as minor and specific as it is.

The export of Roam is either markdown files or a big JSON file. The structure is simple to understand. So I'm not concerned about porting out.

A very interesting read, but I can't shake the feeling that Roam is "just" a better wiki. Which is probably all that is needed to have a product that works, sometimes the littlest changes make for the best tool. Another wiki that is pretty interesting is https://tiddlywiki.com/, although Roam seems to have a easier to use interface.

TiddlyWiki is great, was one of many tools I used before starting to work on Roam.

Lack of easy collaborative option and the speed of organizing sets of intersecting sets of notes were some of main reasons I needed something else.

> organizing sets of intersecting sets of notes

Could you please be more specific ? I'm curious about what this is exactly and how Roam solves it

This clause from the privacy policy is not very encouraging:

> We may terminate or suspend your account and bar access to the Service immediately, without prior notice or liability, under our sole discretion, for any reason whatsoever and without limitation, including but not limited to a breach of the Terms.

I think that's a lost battle now -- the Terms of Use / Privacy Policy of pretty much every online service, important or otherwise, says the same thing. For example, from HN's own terms of use:

> You agree that Y Combinator, in its sole discretion, may suspend or terminate your account (or any part thereof) or use of the Site and remove and discard any content within the Site, for any reason, including, without limitation, for lack of use or if Y Combinator believes that you have violated or acted inconsistently with the letter or spirit of these Terms of Use.

But HN doesn't hold my entire life's writings

Sure, and Github might hold your life's code, but they state exactly the same thing.

> 3. GitHub May Terminate

> GitHub has the right to suspend or terminate your access to all or any part of the Website at any time, with or without cause, with or without notice, effective immediately. GitHub reserves the right to refuse service to anyone for any reason at any time.

https://help.github.com/en/github/site-policy/github-terms-o...

Also see my comment here: https://news.ycombinator.com/item?id=22107783

The battle is not lost. Just don't use this kind of crap. There are alternatives.

I think literally every online service has this sort of clause, and in the absence of language explicitly stating the contrary, a cautious approach would be to assume they mean exactly this. Pragmatically speaking, you should have a local backup of everything you've placed online.

Dev here - open to suggestions

I also don't get how anyone would put up with this.

This might be okay for a free of charge toy, but not for anything that matters at all.

It's not perfect, but you can export your entire database in Markdown format at any time. So given regular backups, this wouldn't be an absolute catastrophe.

Trusting my "personal knowledge" to a proprietary tool seems like a fundamentally bad idea to me, and no amount of clever features is going to fix that.

I believe the founder mentioned that they will eventually allow you to encrypt your data.

If you do any sort of public writing, you could use it for your writing process since that info would be private eventually anyways. Or you could expand that idea and include anything which you would be okay with being public.

For example, I may not have a reason to share a snippet from an article, but I wouldn't be bothered if it became public. I may also not be bothered if notes on that snippet were shared. And maybe if I did some writing for myself for understanding a subject, then that might be fine as well.

That's pretty much how I use the app.

Doesn't it bother you that "the app" will be gone and forgotten in 5 years?

Not if there's no other app which doesn't have the same functionality. I would rather get the power of the app now and then lose that power in 5 years rather than disregard it completely. If the work I did in the app provides an ROI today, then I'm willing to face the possibility of the app evaporating.

I'd like to add to this that since the app lets you export your content, even if/when Roam dies, it would still be possible to move the data to another app somewhat easily.

You can export, but you lose the links and ID's of the sections.

Shouldnt be true for JSON export.

Touché, I haven't tried the JSON export yet. Thanks!

Usually these end with an "Our Incredible Journey" post[0] and 14 days to back up your data, with the site half-broken during that time and nobody manning the support. I'd set up a script to keep your data backed up if the notes are important, or pray you're not on vacation when they get acquired.

[0] https://ourincrediblejourney.tumblr.com

> I believe the founder mentioned that they will eventually allow you to encrypt your data.

Well, as soon as the encryption comes this product will be usable. Giving my personal data to a non open source note taking project without any encryption capabilities is a big no.

Use nvalt till that ships then, import will be easier, and you'll start to taste power of very fast two-way links.

If it were me... I'd probably have the data stored in a postgres db, maybe use graphql+elasticsearch indexes to provide the linking and searching... then the user could simply supply aws creds and what not to their own instances and basically host their own data that way. UI/UX still would be proprietary but that data fully owned by user.

Unfortunately, I tend to agree. I was pretty excited skimming over the post, not realizing this was a hosted solution. Any good self-hosted alternatives with the same features? Or good thread about personnal knowledge management?

Co-founder here.

Best self hosted tools are Org-mode and nValt.

Neither have full feature set, or work collaboratively, but nvalt in particular gives a good 80/20.

We export to plaintext markdown that reads well in either, planning export to org mode with enough data for a good elisp hacker to build parity of main features -- will do sooner if we know there are takers for that.

Have seen cool analysis of JSON export in Wolfram Alpha.

Nobody in their right mind should put anything they can't afford to lose in your service until this clause is adjusted to include some kind of export: https://news.ycombinator.com/item?id=22105480

Good point. Will update.

There's tiddlywiki.

I keep falling into the occasional "I need a note-taking / todo app" rabbit hole, and there is a lot of people / companies re-inventing the wheel, but all of them nowadays seem Electron apps with proprietary cloud storage.

I should go back to a folder of .md files and todo.txt (http://todotxt.org/). Right now I'm using Evernote and Todoist but both of them ship my data off to the cloud - and I have no clue how secure it is. And sure I could do my due diligence but the companies are neither required to, and/or forbidden to reveal who has access to the data.

We store in cloud as it is a collaborative app, designed for building off your own and other's past work.

Lots of self hosted options that are much better than evernote or todoist. Org mode for one.

Worth learning emacs for that.

I disagree as I don't want to spend time learning a third new tool just to use another tool. Especially not Emacs.

IMO "easy" linking to other pages should be doable, but part of Roam's power is creating those pages by just making the link.

The linked references block I think could be extremely powerful in Notion. Bidirectional linking helps to associate data better than anything else I know, and if that can be handled semi-automatically it would blow people's minds.

Also, yeah, opening other pages in a sidebar would be amazing for being able to reference data without context switching and literally losing sight of what you're working on.

Personally, I'm using Notion for structured data, and Roam for unstructured, freely associated data. They're both pretty awesome in their own ways.

> Also, yeah, opening other pages in a sidebar would be amazing for being able to reference data without context switching and literally losing sight of what you're working on.

Can't you already do this? For example, if you right click on a title of a page, you get an option for "open in side bar." Then if you navigate to another page in the main pane, the sidebar still stays at the page you opened there.

You can also shift-click on links to open in them in the sidebar

Or control-shift-O when cursor is hovering over a link to open page or referenced paragraph/subtree (block) in sidebar.

> but part of Roam's power is creating those pages by just making the link

This is exactly how MediaWiki (and clones) have been doing it for years.

Not exactly. Correct me if I'm wrong, but in MediaWiki (and others), you need to click the link, (optionally) type some text, and hit "Save" or the page doesn't actually exist yet.

With Roam, the page is automatically just there when it's mentioned. A subtle difference but I'm finding it quite useful.

In the last few years I repeated the same cycle: register, start of use, enjoy, invite my family, get the announcement that it has been bought, export my data when they announce they shutting down, rinse and repeat.

The first one was google reader. One day it was shut down and however I could exported my feeds, I was following a few people and read some really amazing article at least once a week which I could not find by myself. Also people I was following started to use 3 different platforms each with limited set of funcionality.

I've been using an email client which was bought and killed by google. Online diary app I used is gone now. My todo app was bought and "integrated" into their product by Microsoft. (However, Kudos at the same time, I still can open the doc files I created in 99). Calendar app gone.

About 90% of the mobile apps I bought on my iPhone 3G is not usable anymore (some of my favourite games gone forever).

Even some of the software I bought for my business (and paid double price for the business license) is now gone or not usable because of lack of upgrades (I would have been happy to pay for the upgrade license).

I'm really, really tired of the fact, that no matter if it is a website, a mobile app and from time to time, I need to do another cycle of exporting, finding a new tool for the same task, setup, import, etc.

I'm privileged, because I can setup my own hosted tools for myself and my family, and my current job is also consist a serious amount of "tinkering". I would be seriously be happy to pay if I don't have to do the same cycle again and again and again and wasting my time doing instead of spending my time away from the keyboard. Currently there is no startup which can be trusted that they will be around in a year.

So whenever anybody is asking to invest time and effort to start using their product, which has no clear plan about their revenue streams nothing about how they going to sustain their business is a huge red blinking no. When I hear the expression serial entrepreneur (I throw up in my mouth), it says to me, stay away.

(And lets forget about privacy, because that is another massive can of dragon sized worms)

Edit: business stuff added

The crux of these points is one of the main reasons I’m reluctant to put any amount of time in to new applications.

And, unfortunately, the more of us that feel this way, the less companies get invested in the space, and (it seems) the further we get away from real long-term solutions.

Other than the interface (which is a big "other than"), I'm trying and failing to see how this differs from a managed MediaWiki. Even most of the syntax appears to be similar.

MW would also solve some of the complaints about a lack of mobile support (MW has mobile skins), APIs (MW's is robust, for better and worse), scaling, and aliases (redirects).

Play in the Sandbox on the help db for an hour and see if you still think this.

I did - it's slicker with better UX for sure, but technically what's different?

It looks like the main differences are the bidirectional links and the automated creation of those links.

Bi-directional links are visible if you go to the What Links Here [0] section of a page. Notably though it doesn't include "unlinked references" (as Roam refers to it), which is quite useful.

[0]: https://www.mediawiki.org/wiki/Help:What_links_here

You can still fish that up if you need to either from the UI's search or the search API, but in poking Roam that felt like a potential anti-feature around ambiguous terms, for instance with no clear way to flag a particular reference as not relevant to a term.

Blocks (paragraphs with subtrees) as first class objects with uids that can be pointed to.

Pages as filterable collections of backlinks from blocks.

A datalog engine for structured data via attributes.

Collaborative real time editing.

But you're mostly right - UX that comes from that is the thing that matters.

OK. I have. I do still think this. The interface differences — which I acknowledge, and are significant, and maybe that's worth the switch — haven't landed that way for me yet. But I just do not see any differences in functionality, and I'd have to give up _a lot_ to get the interface.

The live-edit interface is double-edged. It's a split-second more convenient to edit but easier to make mistakes and harder to revert anything more than a typo. Every time I miss a TODO I end up editing the TODO instead and breaking it.

There's nothing in here that's appreciably better than MW's raw editor. I can't enable multi-cursor editing in Roam, and I doubt it would ever be an option across nodes/list items.

If I run the window vertically (or especially push to 1/3 screen width) I lose all value from the right sidebar.

There seems to be a lot of functionality tucked into right-click menus that are inconsistent, _extremely_ context-sensitive, and in some cases require considerable dexterity that I don't have. (Right click the _bullet_ for formatting?) Why aren't they keyboard shortcuts for these functions?

Block editing and versioning are neat, but less powerful than MW templates — again, a feature for me, probably not for the target audience? It's easy to create new versions in blocks — too easy, because how do I remove one I've added by accident? Are whole pages versioned?

The inline reverse mentions/backlinks are nice, though I don't know what they're good for yet. I already easily take advantage of backlinks in MW without needing them cluttering the inline content, and I can already sketch out how to add them in MW using its API if I can better grok the value of surfacing them. Otherwise they're already — at most — a click and keyboard shortcut away.

Surfacing full-text search results in the live search bar drags the interface down immensely, and isn't what I want out of that bar _at all_. That's the only thing I can see that's net negative. Browser lag on the third letter of a four-letter word was measurable in seconds — just chill out and let me enter the word.

The on-page context filters are neat. That's a feature that would legit take some work to implement in MW.

Bottom line, if the hook is "embeddable backlinked content in a lightweight editor", I get that in a non-VisualEditor MediaWiki install, plus a superior text editor and an API. And the API — again, to me, acknowledging that if I don't see the value Roam adds to that model, then I guess I'm not the audience — is still by itself a total dealbreaker even before I get to not being _able_ to self-host, not seeing the backup/restore story, not having any offline editing options, and not having a useful mobile editing story.

Might not be for you.

Most of benefit is in the new workflows you get via backlinks.

Lots of good feedback in their comment.

You told them to try it. They did, and wrote up feedback. Your reply, after you told them to do it, is dismissive.

A good deal of the feedback remains applicable even if you’re ‘sold’ on the benefit.

question for anyone who may be interested - what would it take to implement those workflows with backlinks in MediaWiki - or is it just outright impossible due to current data structures used.

Off the top of my head, probably the Backlinks API[1].

There's also ongoing experimental work toward making the parser's backlinks support more usable.[2]

[1]: https://www.mediawiki.org/wiki/API:Backlinks

[2]: https://www.mediawiki.org/wiki/Extension:AdvancedBacklinks

This looks an awful lot like nvALT (https://brettterpstra.com/projects/nvalt/) or Vim-Wiki or any Wiki software for that matter.

It can be easily replaced by MediaWiki for the backend but to it's benefit the UX is very good and is a very good choice for people who don't want to muck around with MediaWiki.

But given the current state I cannot use it long term due to non-availability of non-desktop use, no encryption and no good PRICING or an idea of pricing.

Having said that I'm keeping this in my radar and shall revisit periodically.

Trillium Notes have been mentioned before [0]. Provides similar functionality with a lot more power.

It's open source and self hostable.

It's possible to script additional functionality into it from the UI itself. I find it quite powerful.

[0] https://github.com/zadam/trilium

This looks awesome! I've been wanting a personal knowledge store for a long time. Between bookmarks, random notes and lists that I throw away, cool quotes I find, and even thoughts that I ponder and get distracted from, I feel like there's so much interesting stuff I've found that remains inaccessible or was lost or is just laying around uncatalogued. It's just that all the usual data-cataloging systems seem so clunky or have a whole "system" that intrusively proscribes how to use it.

Roam looks like a pretty interesting solution to this problem. Simple but capable flat document system, especially the very nice tagging with automatic bidirectional linking to organize everything with. Pretty UI, maybe a little more customized than my tastes, but still nice to interact with.

My only reservation is the reason why I haven't tried any of the other big proprietary solutions: I really don't want to switch away once I'm on it, and I don't want to ever worry about losing access to it. I.e. I'll want to self host it. Don't get me wrong, I would still be quite willing to pay for such an app, but I also need it to be open-sourced and have a simple open data model.

To that end, it seems like this UI on top of Perkeep's API https://perkeep.org/doc/overview might be pretty nice combination.

He doesn't seem to understand tags and links in evernote. Evernote can absolutely be structured as a flat graph in the way that he describes as a big benefit of Roam.

That said, as a long-term evernote user, I've been wanting to leave for a while and move to an OSS/self-hosted solution. It might be joplin but my few attempts to use it I haven't yet had enough time to evaluate whether it would be a viable long-term solution.

Edit: I'm also seriously considering just using https://github.com/Alok/notational-fzf-vim and only doing note-taking on my laptop (not on mobile). It's extremely fast and basically zero friction. This would have a slight downside in that I quite often want to consume notes on the go so I may need to find an option there (eg serve my notes up using a private server or something).

Problem w evernote is that file is bad unit of organization.

In a meeting with 4 people, discussing three big ideas, related to two projects, what file do you put the note in?

In Roam that's not even a question you have to ask, all info in each paragraph goes to all the right places

This is definitely a problem I run into. I’m having a team meeting about Project X, and someone mentions that a particular task is waiting for something from Project Y.

The only way I’ve found to handle this is manual. Note the dependency in the meeting notes, then process the notes after the meeting creating relevant todos elsewhere. But that’s only maintainable if you don’t have a lot of work on at the same time, or a lot of dependencies - and we find ourselves back at lean.

Too much work is too much work, and no amount of automation is going to fix that. Furthermore, life is messy, and there is no universal way of structuring that mess. The more mess there is, the harder it will be to live with it. So reduce the mess to the point you can live with it, rather than trying to structure it.

I don't use files, I use tags. I have 2 notebooks to differentiate notes that I want synced to my mobile devices, but really that's not needed.

You can tag any note with multiple tags and then use saved filters to always find notes which are in a given tag. All of this used to work a lot better (evernote search has been getting consistently worse) but it still works fine.

You can use multiple tags, but the granularity is still page level. To continue on Conor's example, you'd have, say, a note for your whole meeting, with 3 sections for each agenda point. And you'd have 3 tags.

But then when you want to see information pertaining to a given tag, it means that the content on that meeting note is 66% irrelevant. That doesn't happen in Roam.

But in the current state of the app sharing a note shares your entire DB, its an awesome tool but I'm unable to find a way to create a secondary DB without creating a new account.

Capability is there, but UX sucks for both sharing single pages and creating seperate workspaces.

Been working on both this week.

could you point me to the doco

https://roamresearch.com/#/secretinvite

Have you considered vimwiki http://vimwiki.github.io/?

I use it all the time. My vimwiki dir is the main thing that I search using notational-fzf - they work great together.

That looks like an amazing combination. Do you have any other plug-ins that augment vimwiki?

For one, taskwiki, which if you use task means you have a great way to bang a checkbox in a vimwiki file and have that create a task.

MY company uses outlook so I sync the outlook calendar to khal using vdirsyncer. I then can use a (really janky) vimscript function of my own devising to add the calendar and some standard stuff to a "daily note". So I go <leader>ww and it creates a note for the day and adds my calendar and then I use that for all the basic notes for the day.

vimoutliner for when I absolutely want an actual outliner to do some planning. It's not always the right tool, but when it is, it's great, and it integrates perfectly with fzf. The other thing is that because notational-fzf is just doing a fuzzy find over text, I point it at all my sourcecode dirs for immediate jumping to particular files, functions etc in sourcecode.

No I do, they just don't offer this same level of functionality. Try the tagging system in Roam for a few hours and you'll see the difference, Evernote's tag setup is much clunkier and harder to do on-the-fly.

This actually looks very interesting! I have been trying to get all of my immediate family to coordinate on a shared google document to try and capture a lot of our family history, but the unstructured nature of memory snippets and anecdotes is challenging to organize coherently.

For example, I want to be able to easily capture the fact that we owned a dog. Oh wait, but what about all the other dogs we owned over the years? What about the time I got bit and had to get stitches? what about the other times I had to get stitches? What about the times that other family members got stitches? I find that memories oftentimes don't bubble to the surface until they are sparked by other memories.

I want to be able to easily organize everything in a connected graph, but also still map it to a timeline.

I'm definitely going to have to give Roam a try. I wonder how well it works when it comes to collaboration between different people?

I've never managed to find Roam's pricing. As most apps have a link to pricing right on the landing page, I take it as a bad sign when I don't see it.

Founder mentions on twitter pricing tbd though hopefully this year, considering $30/mo with a discounted annual plan. Discounts/scholarships for students, non-profits, and low income. Also mentions a $10k lifetime plan with a potential “code escrow” should they cease to operate (though acknowledges not expecting to get many of those—mentions expecting 10, ever).

$30/mo seems insanely expensive for a note taking app. I guess I'll stick to the local vscode on top of a notes directory that takes advantage of local access and indexing and runs on a synced folder.

It's insanely expensive if you are comparing the price and functionality to existing apps. I would expect a "me too" app to have "me too" pricing.

The reason people are talking about Roam is because it's not like other note taking apps. People who use the app will need to extract a break-even $30 / month worth of ROI just like you might from any other piece of equipment or tool which extends your capabilities.

I find Roam compelling enough that I believe I can reach that point. If I'm not reaching that point, then I need to adjust my efforts.

I feel pricing shouldn't be a sticking point. I try to keep my consumption to a bare minimum, so that my focus is on high value activities. If the tooling can't return the investment, then that activity and the related tools aren't for me. It's not that the tooling is bad, it's just that my efforts are best pointed elsewhere.

I believe the founders operate on this idea and it's refreshing. Yes, it will be expensive for many. Potential users will walk based on the pricing. And I wish more developers would be willing to create something different and charge proper pricing for the effort.

Despite what the article claims, I really don't see much new in what Roam does over any other note taking app or wiki. --Whether they have more or less structure, they all let you link pages together, so hierarchies are generally up to you to decide if you want more or less of.

I use OneNote since I already have O365 anyway. To convince me that I should pay more than 3 times what O365 costs, you'd have to be able to show amazing improvements, and Roam just doesn't.

Seems similar to Mindforger ... Mindforger I think uses ai or something to 'link' internally between docs. Just found it the other day, but it's free. so -- I'll probably stick w/ that.

> I believe the founders operate on this idea and it's refreshing.

Many founders do, and that's why they price themselves out of the market. That's simply not how people make purchase decisions. There are many, many products out there that are "worth more" than the price. If you think you can charge $30/month for a product because your customers will get $31/month worth of value from it, you won't have many customers.

I understand people being disappointed with the numbers you quote, which I agree seem very steep, but I have no idea why they downvote you, for providing the seemingly best answer (currently) available to GPs issue.

It seems that right now it's considered beta, so all accounts are free trials.

I can't tell where "Roam Research" is based, or if it's even an actual company or just some developers using the name. There's no terms of service beyond their privacy policy, which doesn't list what jurisdiction they claim to be under.

I definitely wouldn't risk using the service based on the lack of information they provide.

Hi, I wrote this! Thanks for sharing yarapavan.

Happy to answer any questions about how I'm using Roam.

Minor nitpick. His whole description about how "fluid information" works and how it's so much better than rigid/siloed hierarchy:

Really it's tree-based vs tag-based organization [1][2].

Tag-based organization turns the collection into a graph. Simple as that. IMO it's the superior way, since a graph is kinda like a superset of a tree, but there is nothing novel that was done by Roam here. You could have tag-based file systems too [3].

Note: when you have tags, it has to be capable of multiple-tags (not just one), otherwise it's not much different from a tree structure.

[1] https://www.nayuki.io/page/designing-better-file-organizatio...

[2] https://zapier.com/blog/how-to-use-tags-and-labels/

[3] http://web.mit.edu/6.033/2011/wwwdocs/writing-samples/sbezek...

Actually going to look into this - given up on Evernote, don't like Apple Notes and Bear writer just didn't click....but did anyone else get put off by the whole "casually going to show off my notes on Prepping and Bugging Out as they're totally normal and chill things to be across". Undermines the credibility of the writer.

I thought it added to the credibility as it positioned them in my mind as a longer term thinker than myself.

By the way, I’ve been bouncing between bear and Ulysses and I’ve decided that I just love the Bear experience more than any other editor. There’s something really elegant about it....

It's crazy to me that someone could have so many notes that they need to do all this tagging just to be able to find them. After 2000+ notes in Apple notes I've learned that the notes I go back to most often are nonfiction book notes and certain periodically updated lists, and any other notes I just find through the search bar.

I've been through a hundred note apps, in the end, taking notes is for the brain to help remember. The app is your brain. Generally now, I through stuff in Google Docs and move on. It's searchable, good enough.

can't store work confidential data on it because its not a local app.

At first glance it looks somewhat like MediaWiki (the wiki software behind Wikipedia), which also supports creating pages implicitly by linking to them. The markup style also looks similar. Does anybody know how it compares? The MediaWiki interface looks clunkier and you probably will need extensions for some of the stuff Roam does (e.g. like Semantic MediaWiki).

Have been using Roam for a while now. Very powerful potential. I’m excited to start a second database and use it as a personal CRM

I wrote about this idea ("Zettelkasten") on our blog. Here is the link in case you missed it:

https://emvi.com/blog/luhmanns-zettelkasten-a-productivity-t...

Stupid question - but I can't tell if this can be used by multiple people as part of a company KB. For me (and for many) sharing knowledge within a company is a bigger win than my personal KB (which I happen to use workflowy for). The perk of this it seems to support images & videos which workflowy does not.

It can! You can share your database with people and all collaborate on it. I've been debating doing that for my team to replace our Notion wiki.

Is there a "team" or "company" specific version? Or is it like workflowy where you would have to explicitly share at different nodes.

I made the switch to Roam for a week and put my thoughts into a new post on comparing to Notion.

Notion vs. Roam: Which Tool For Productivity

https://www.austinrileygray.com/articles/notion-vs-roam

This has me really intrigued and I've been on to workflowy and a smattering of org-mode for over a decade now. Consistency is more important to me than optimality and workflowy has served me well for that. Reading on interestedly ..

I worry for my data. If I invest a lot of time to build a repository of information, how can I archive it, extract it out of the platform. It is in early release, and they seem to avoid the question about pricing.

What data structure would be best suited to implementing this set of features?

You can't even sign up for Roam if you use a pihole for DNS, since it redirects you to a blackholed site. Okay. A good reminder of the commercial interests I don't want in control of my notes.

I just read about this and find it interesting. Two questions:

1. Is Roam text only? I haven't noticed any images in the examples I have seen so far.

2. What is its business model? I didn't see any pricing information.

1. Yes, you can upload images.

2. The co-founder has said he's looking at charging $30 / month when it's ready.

https://twitter.com/Conaw/status/1214867422496165888

This looks like a gui vimwiki.

Does anybody know what Roam's business model is?

i used 'tomboy' a while ago which has support for links between notes

So do lots of other apps. You can link between files. Even just plain markdown would allow you to link between files. You could just include a build step and edit MD files and read HTML.

But ROAM also allows you to reference sections. Links you create leave references at the bottom to show you which sections link to this particular page or section. That gets to be a slog with plain text.

I have that problem with many sites that use that. It's usually because of privacy blocking so I just use the regular login.

so popular, i love it

TiddlyWiki has many of these auto tagging and page creation features. I enjoyed using it for a time. It's a self-contained portable file.

Its homepage is an excellent demo of the software itself: https://tiddlywiki.com/

Reply on news.ycombinator.com