Since when does one need an “excuse” for markdown?
Using markdown is a choice, not an excuse, and that choice should be based on some facts that are interesting for the (potential) user, and not for others.
Those that do not want to use markdown don’t need an excuse for rejection either, it should be based on facts.
On top of that, both groups may have a subjective, non-fact-based (partial) reason for their choice.
There is I think a key difference between a wiki and Obsidian. In a wiki you can indicate a link that should exist within the wiki, but might not, while writing a page. In Obsidian you only link to existing material while writing a note. The former often creates a load of work ‘to be done’ (like the red links in Wikipedia), whereas linking to existing material only creates a network of content to explore, and no ‘missing’ material.
Wrong. In Obsidian you can put any word or words between double square brackets, which, of you click on that, will create a new note that’s linked to where you came from.
While I appreciate why Markdown is a positive, it’s very hard to agree that a Rich Text editor or WYSIWYG would be a barrier because of more options to fiddle with. If anything, having to write in Markdown and then see a preview to ensure you’ve not missed anything is a barrier that some people are going to run into right away with Obsidian. Again, I’m not making any argument against all the good reasons, but I cannot see any way writing without any friction in styling or format won’t be a positive for everyone. The semi-invisibility of this in Roam is a big plus.
And it’s even one step better than the typical red MediaWiki link, because Obsidian is aware that the missing link should exist, it will autocomplete it for you in other places as well, which is a feature a use a ton now.
You could argue that it’s a matter of personal preference. I’d actually challenge that, though.
All software should optimize cognitive load for the task at hand. There are generally three kinds of cognitive load: intrinsic (what must your mind—and your working memory—literally do to navigate the conceptual elements of the task?), extrinsic (engaging with the material elements, like using tools, required to execute on the task), and germane (the thinking required to link the intrinsic effort of the task with whatever resources you might have already).
In general, the user should be able to put as much of their available cognitive resources into dealing with the intrinsic load as possible. Doing so is better for engagement, learning, and more productive thinking on the task.
This means minimizing extrinsic cognitive load by making it as easy as possible for the user to do whatever it is they’re trying to do with the tools they have.
I differentiate between the task of styling the work from the task of creating the work. Thinking and writing is different from look and feel. There are overlaps, but at the early stages of writing, it is better to make sure you’re saying and structuring what you mean as effectively as possible before you style it.
Combining these two tasks increases extrinsic cognitive load. Ergo, the more tools we provide for layout/styling for the user when they should be writing, the less energy they’ll have to put towards thinking. This is especially true when the use of styling features requires a mouse-navigable GUI. The goal should be reducing the number of interface elements the user has to interact with to communicate their meaning—asking them to stop looking at their writing, glance around for toolbars or whatever, move their hands from keyboard to mouse, and to go click a menu or button is incredibly destructive when you do it a thousand times per document.
In turn, I think markdown is the best tool we have for WYSIWYM—what you see is what you mean. To challenge @Dor’s point, plaintext does not suffice for this. Structure is an important part of thinking. Structuring writing is different from styling writing. Markdown’s inline structure provides low-effort affordances for structure. (Aside: Adam Hyde’s article is a cheap critique. It uses handpicked examples—like using asterisks for everything instead of the very common _italicization with underscores_ and **bold with asterisks**—to argue its point, and doesn’t really offer an effective alternative in place of markdown.)
The research on this is clear, to me. Here’s an example—“Towards Reducing Cognitive Load and Enhancing Usability through a Reduced Graphical User Interface for a Dynamic Geometry System: An Experimental Study”:
… we designed the interfaces and carried out an experiment involving 69 undergraduate students. The experimental results indicate that an interface that hides advanced and extraneous features helps novice users to perform slightly better than novice users using a complete interface. After receiving proper training, however, a complete interface makes users more productive than a reduced interface.
It also shows some of the nuance at the centre of this subject. The authors are talking about a “dynamic geometry system” in which visual elements are a key component of the task, and in which certain elements of the task greatly benefit from visual UI elements. I think you can generalize the lesson from that to Markdown to imply that people think more effectively with less UI, and training them on how to use a optimally-minimal UI is the best case scenario for maximizing effective thinking.
One last note on Markdown: I suspect that the need for previewing is an artifact being used to WYSIWYG editors like Word. When I write, I don’t use previews. Ever. Markdown’s inline styling makes it easy to “preview” the structure of your work without needing to actually change modes. I just think folks need to get used to it (see the note on training in the paragraph immediately above.)
Of course, all of this is foolish if we don’t delineate what we mean by a WYSIWYG editor. Editors that limit style choice while providing highly accessible affordances to engage in structuring are probably just as effective as markdown. The CodeMirror (or CodeMirror-like) WYSIWYG editor that Obsidian will eventually use is pretty close to this, although I still worry it will provide too many options to optimize cognitive load.
My specific critiques are against apps that combine layout and editing, like Word. Markdown is optimal because it minimizes input extraneous cognitive load, but simple WYSIWYG editors are basically equivalent otherwise.
My bad. Hadn’t encountered that myself, as the flow of autocomplete while typing a link suggests picking from the existing links.
" aware that the missing link should exist, it will autocomplete it for you in other places as well,"
That sounds very very useful.
Is there then also a way to list non-existent links?
AFAIK, there is not. But what you can do is look at the Graph and you’ll see notes that are not linked to any other notes, they’re just drifting alone in space, like a lost satellite.
That doesn’t exactly help if you’ve entered a link to a missing page, I think that happens is that the missing page ends up on the graph, but it’s linked to the node you wrote the link from, so it won’t be isolated and easy to spot. On the other hand, maybe styling missing nodes on the graph would be related to this: CSS support for "missing pages"
Thanks all, for the replies. Very interesting for me the two main discussions:
Backlinks seem to be a tool to make it easier to link pages one with another and avoid lost pages. So it fits with the idea of the wiki, helping to tighten the relation between pages. Transclusions, on the other hand, seem to me very new and interesting. I’ll have to test it. It would mean that you need to structure your note in a way that a paragraph would make sense in another note as a transclusion. Sometimes paragraphs in a note flow one into another, so it might be difficult to transclude it to another flow of paragraphs. However, sometimes a paragraph might just be a self-contained piece of information, so I guess in this case, it could be used in different notes.
As for Word, we use it heavily where I work and it does the job. We have to produce documents collaboratively between 30-40 pages every month, with a 6-level structure (headings).The WISIWIG is important as we try to make this boring document attractive with maps, tables and blocquotes… Also the document is processed with VBA (Visual Basic) so that we can reuse the content in a database. That where lies the weakness of Word documents for me, and why I came to Obsidian in the first place: they are neatly stored in some folder and forgotten, with almost no way to reuse the information inside. The database transformation does the trick, it’s not very sophisticated (it goes to Excel), but we can filter in various ways.
Now let’s imagine that the document, or better the 24 such 30-page documents produced in the last 2 years, are in Obsidian instead of in a folder. It would create a kind of Wiki with the possibility to create additional thematic pages including transclusions. For example, pages could be created for a number of keywords that are scattered in the 24 documents. We would get two years of knowledge that could be analysed and reused. I am in that 
I have to see if it works.
Thanks again to all,
As a contribution I’ll make a separate post on TheBrain. As said by arcandio, it’s proprietary so I’m stuck with it the rest of my life, hoping that they will not close some day 
But they have been around for more than 20 years (slow-paced development though…), and remains a fascinating tool that I use every day.
Here are a few screenshots of how TheBrain was used reading this thread and trying to understant how Obsidian was more than a wiki. Screenshots makes it small, I hope you will be able to see.
I created the Obsidian node about two weeks ago, while I stumbled upon Roam:
Now, if I click on Mardown, I get this:
And if I click on Take Notes, I get this:
And so on…
The graphic visualisation makes all the difference in this case because the focus is not so much on linear text but rather on linking concepts and creating some order that makes sense to me.
For each of the nodes, you can add a note, and the note accept hyperlinks to other nodes. So, TheBrain could also be used as some kind of wiki. But Notes are not as sophisticated as in Obsidian/Roam (no backlinks, transclusion, outline, basic search engine, etc) and they remain kind of lost in the graph. So now you know why I am here
To answer a couple of the orange questions:
BTW, Roam Highlighter has been adapted to work with Obsidian too.
I used The Brain in the early '00s as my desktop (both for notes, as well as navigating files)
Have been familiar with the Brain since its very earliest incarnation. Over the last 10 years+ I’ve paid for an annual license/subscription 3 times and tried each time to make it my second brain. The last one thinking “this time I’ll make it work” but it just doesn’t get there for me. Their incorporation of linking from mind maps and making it visual was visionary but the system is just cumbersome. It looks awesome with its visual rearrangement when you move through the knowledge base but for me in usage it was just clunky. Having to move a “thought” (their term for a unit of knowledge) into the center of the plex before you can see its note or other attachments is just too high friction. It’s too bad because if I could have gotten it to work I’d have many years of second-brain accumulated in one place rather than scattered across dozens of apps or notebooks as it is now. So now you know why I am here 
What will happen to all those “thoughts” you collected over the years?
That’s a good question. Plan is to bring those most currently valuable into Obsidian if it gets to where I believe it will be “the tool”. Many are in paper notebooks on a shelf, many are scattered in apps - those in apps are more vulnerable to loss so slightly higher on the priority list. Prioritization being also a function of current value.
Understand I am an old fart but I was lucky to discover early in life that in order to learn something you have to think about how best to organize the information and WRITE down your conclusions. As a result I have a lot of notebooks that are rarely revisited - the main value was the process of taking the notes and reviewing them at the time. I doubt those will be captured (it would be difficult) but it’s not a big loss - I learned the information in there. FYI my observation even now with these digital tools is that I get far more value from handwriting notes as I read then thinking about how to organize them THEN taking them into digital form than from those books where I use the various Kindle capture methodologies. A big portion of the value of note-taking and zettleing is in that step of thinking about how the knowledge fits - and that is done more deeply while handwriting.
Also when will I know Obsidian is THE tool? When there is a step-change improvement in filters/queries. Until that happens it is not the tool I am ready to commit the time to import.
There are those who are experts compared to me the ignorant, and they think that hand-writing is more effective to remember your notes. So, by all means, go ahead if it works for you.
By "old “thoughts” I meant the notes in The Brain. And the reason I asked about them is that over the past 7 years I have had to change computer and note-taking app a few times, and each time I ran into that wall of vendor lock-in.
So, I decided enough is enough, I need future-proof notes that are mine, so I opted for markdown.
I am still in the process of manually transcribing (copy/pasting) my OneNote notes to Obs.
By "old “thoughts” I meant the notes in
The Brain.
Ah I see - I never had anything in the Brain that I also did not have elsewhere (mostly in OneNote to be honest) so it was not an issue. When I last used it however it was not possible to do any kind of bulk export so pulling information out of it was a “one at a time” process.
TheBrain allows for 3 types of exports: folder hierachy, text file, JSON files. I have never used any of them so I wouldn’t know what result you get, and frankly I hope I will never have to because I am kind of expecting a mess. They say JSON files would also export the links, which would be the biggest loss in case of leaving TheBrain: attached documents and notes can be exported, but I am not clear how the links are exported and could be reused in a different software. I don’t know how JSON files work.
In any case, even without exporting, TheBrain keeps notes, documents and links in folders that can easily be opened and managed with a file manager. Folders are coded so you wouldn’t know to which node they correspond. Notes are saved in html so easily readable, but they are named “notes.html”, so impossible to know what they are about before opening. Maybe exporting as a “folder hiearchy” would rename those notes with the name of the node, who knows.
I suspect the issue of exporting links is common to any software that produce links between nodes. As far as I know there is no industry standard for this so far (xml? JSON?). So in the end, once made a choice better stick to it, and hope for the best… To me, the accumulation of data in a single system has outweighted so far the inconveniences and imperfections about TheBrain. I am now trying to see if I can somehow complement with a smart note software.
