One lesser spotted feature of Luhmann's system is his advice to work on anything only for as long as you want and then switch to something else.

Since starting to move stuff into Obsidian,  I've noticed that this has really improved productivity. It's because I now have one system and one workflow for everything. I never have to switch from one set of toys (progs and files) to another. Writing,  research: just do it - project switching is all in the brain. Think about dinner,  need recipe is all the same. Temperature too high? Ditto.

It's not the single boundaried academic system envisaged by Luhmann,  but my brain's never been great at boundaries and this allows me to follow its lead, with no friction. And I don't have to be uncomfortably disciplined beyond what comes naturally.

Not so smooth on Android, but I remain hopeful that will be sorted around the end of the year. 

I can understand that.

I find using Vivaldi helps.
That's what I tend to use for new stuff.

And editing old can be done on quite a few programs.

But I'm looking forward to having a native Obsidian app. I'd be cogitating more if I didn't know it was coming. Or if the developers weren't so fast.

I think that makes sense for anyone who has VSCode open all the time.

But for someone who doesn't, it's the reverse.

The main programs I always have open now are Obsidian, WriteMonkey 3 and Vivaldi.

And nothing I'd call a zettel.

A Beginner’s Guide to Note-Taking (for your Life):
https://www.ivan-ang.com/a-beginners-guide-to-note-taking-for-your-life/

Atomic notes: How to use Zettelkasten to boost your creativity and productivity:
https://zapier.com/blog/zettelkasten-method/

μPad - a powerful note-taking app that helps you organise + take notes without restrictions

Roam Monkey:
https://roamresearch.com/#/app/roamhacker/page/jI-X_cwaf

Here is the maddening dichotomy of Roam, Roamcult, and Conor laid bare: Things like this - which, by the way, looks really useful - are a testament to the flexibility of the system, but also an indictment of the development priorities of the team lead.

It is on the one hand nice that Roam is flexible and extensible enough that such things can be done within it, despite them not having access to the code. That is really cool, in a way. It allows the community to help cover things that the core devs cannot, and without open sourcing, since not all developers/companies and profit models are amenable to doing so. It does present some security risks to allow this kind of stuff though, even with the warnings.

On the other hand, the security risks are arguably the smaller concern here. A majority of the features in RoamMonkey are highly useful to most people. They are, in fact, largely features that should be a core part of Roam! Maybe they will be some day, but here's what frustrates me so much. These features were achievable by a self-professed amateur JavaScript developer using only external access. Surely this could be done as well or better and faster by the internal developers. And yet, in the same several months that this guy has been developing these highly useful features that almost everyone would use, Conor and his partner(s) have added a Pomodoro timer, Mermaid diagrams, and other arguably niche and certainly not "first priority" stuff. This is behavior that users SHOULD NOT accept from a developer of an app they are paying for (and, I would add, paying a premium over many comparable products).

Roam is doing cool stuff, but I refuse to support Conor's self-indulgent approach to development priorities. It's not going to work out well in the long run.

- Oshyan

The boom in interest in both Roam and Obsidian, and various third-party stuff, is both good and bad. Good in that great things will probably come out of it. But bad in that things are still so much up in the air that we don't know how much these tools will change 6 months from now. So I still find it hard commit to either ATM, since which one (or other) I will prefer in the not so distant future probably depends on features not yet there or not yet refined. I'm kind of stuck in test drive mode

The boom in interest in both Roam and Obsidian, and various third-party stuff, is both good and bad ...
which one (or other) I will prefer in the not so distant future probably depends on features not yet there or not yet refined. I'm kind of stuck in test drive mode

I don't think of Obsidian as a commitment at all. I see the files as a commitment - and my system of creating and managing them. I'm happy to use other programs in addition to or instead of.

Which leaves Roam and other database programs. And there I am waiting to see. Either they offer a sufficient incentive for me to use them or they don't. I considered believing (despite never having used it) realising that the first year at least would be a write-off; I glimpsed a pot of gold but surrounded by too many red flags to take the risk.

But there's no loss. I don't have to wait where I am,  I can go forward knowing I can change path in the future if I so decide.

That's an interesting take on the situation.
I share concerns about the professionalism of Roam's development. The offline enticement to believers through a PWA, risking data loss if cache is deleted before sync with the online database, just seemed the wrong way for the program to work.
But I hope it does well because a database has advantages that files don't - even if I prefer files for my own use.

Are the advantages you're talking about really a distinction between "files" and "databases", or are they separate? I would suggest that in fact you can probably do all or most of what Roam does with "files", or at the least by adding a database *in addition to* the files (i.e. a database that manages the files/interconnections). Which is to say I don't think Roam has any kind of monopoly technologically on the benefits it purports, and Obsidian could replicate all or most with the will and the time put into dev. Correct me if I'm wrong though.

One is that he's just substantially expanded the team; I have no experience of development teams, but in other areas it's not unusual for it to take time to become productive rather than a drain on existing resources.

Is there any evidence that he did, aside from the customer relations person I think he mentioned?

Hopefully they predicted this and had stuff like the pomodoro already up their sleeves to give an illusion of movement.

Does that really track with your knowledge of how Conor has been operating though? That he would somehow develop something in advance but not release it just so he could then release it later to make it seem like there is progress happening during an otherwise slow period? Doesn't sound like it matches with what I've seen, at any rate.

The second is that a huge number of cultists are desperate to fiddle personally with the program. Mostly, it seems to me, because they have a drive to fiddle (enhanced by lockdowns) rather than a particular need.

This is true and I have no problem with them doing so, in fact I think it's great. What I was pointing out here was that the RoamMonkey features were, for the most part, *really effing useful* and *should be in the core*. In fact Roam added a new feature a mere few days after that RoamMonkey vid was released which does something similar to the Template feature. It is more powerful than RoamMonkey templates, but unsurprisingly (because it's Roam and Conor) it's *harder and slower to use*. Anyway my point is that this is not someone fiddling just to fiddle, this is a smart person (RoamMonkey author) seeing *important* things missing from Roam and spending his time *as an amateur* to develop them externally.

This keeps them attached. Other programs like Trello have benefited hugely from third party enhancement. And I notice the building excitement in Obsidian over the near-term API release.

Trello has an API, as you mention. Obsidian will also have an API. These are reasonable programs to develop for because they give you an appropriate channel to do so. Roam does not, yet. It lets you embed live-read CSS *and* javascript in its fricking pages! That's an insane risk, if I know anything about web tech security. And judging by my own experiences of massive slowdown in experimenting with CSS theming in Roam while the system tried to interpret what I was writing apparently in realtime, I think it's a crappy way to extend a system anyway. Even the RoamMonkey dev admits the fragility of some of the things he's built.

So it's great people have the enthusiasm and interest to build stuff for Roam. But I *don't* agree with Roam dev's decision to allow live interpretation of web languages within Roam DBs largely to enable this kind of hackery. I get that they probably wanted to encourage hackery and didn't want to have to wait for - or spend time on - a proper API at this stage. But I think it's irresponsible. I think there are a huge number of "cultists desperate to fiddle with Roam" because Conor is encouraging that, but I don't think he's doing so in a good way for a professional app.

Let me say one last thing which hopefully clarifies where I'm coming from. If this were a free - and especially open source - app; if this were something people could host on their own machines and expose themselves to the risks by individual choice; then I would have less concern with the "lack of professionalism" that I think Conor is exhibiting. But no, they're charging what is fairly clearly a premium for a hacky, messy system that, yes, is super cool, but also is developed largely at the whim of a potential egomaniac, and the community and his relationship to it is just feeding that. IMHO of course.

I get why people are excited to be a part of it, that very messiness has some exciting aspects. But people who have lost data have a more realistic view of it, and there is the potential that more data will be lost because sloppy devs and irresponsible management lead to things like that. If that does happen, hopefully people will see that it was questionable to put that much faith into the "roam cult" and Roam and its founders, because at the end of the day people are excited about doing real work, better work, and to do that you ideally need a reliable tool. Cool, new, and exciting only get you so far.

I'll take Obsidian's humble, experienced team approach any day.

- Oshyan

Obsidian can't export to markdown because it doesn't contain the notes in the first place. The files have an independent existence and can be edited using any program at any time.

Everything written using Roam is contained in its database. It can be exported, but then it's no longer in Roam.

Obsidian's linking and transclusions can be duplicated at any point by programs that can interpret the same syntax and have similar functionality.

PS Roam's exports have to converted to be read in Obsidian.

How to Take Smart Notes: A Step-by-Step Guide:
https://www.nateliason.com/blog/smart-notes