Hacker News Reader

Comment by slightwinder | Hacker News Reader

The first sentence of the documentation already says it: "turn any set of notes into a powerful database". It's really just that, basically. It's a database-view, where the vault is the database and the rows are your files. There is a fancy GUI for creating views, and it seems there is the ability for live-editing data from within the view. Basically a more user-friendly replacement for the very popular dataview-plugin.

Maybe it's a bit harder to understand, as it's a more mushy than the usual relational database.

View on HN

> The first sentence of the documentation already says it: "turn any set of notes into a powerful database

No, horrible job at explaining. What does it mean to turn any set of notes into a powerful database? What does it mean to "turn"? Does it mean that a file will become a database? Or does it mean that a file can be interpreted as a database? And why set of notes? If I have a single note, can I turn that into a database too? Are the records of the database files, or items in a file? What is happening when I type ![[Untitled.base]]? Is the file where I typed that a database now? Or does that text assume that the file named Untitled must be a database?

They do a horrible job at explaining it.

I'd like a demonstration of what's powerful about it.

Like, why would I want to take advantage of this, and how? I'm with you here. I don't get it. I can stick data into SQLite and do all kinds of neat stuff. Why am I preferring a mushy database trapped inside Obsidian?

This is obviously for people for whom notes are a primary knowledge storage mechanism, and for whom that mechanism is obsidian.

I “could” stick data in SQLite but if I’m not then it’s quite useless to me.

Well everything comes in shades. Lets say you write meetingnprotocols of your meetings down with obsidian. The markdown/freeform and cross-linkable text body of obsidian is perfect for this. But you'd now be able to have a bit richer search over the notes based on who was present. This allows relatively normal people to do that, without having to dive too deep into databases plus it already lives in the app where you take the notes.

As someone who uses Sqlite for a ton of things I don't think the exietence of sqlite makes this useless.

> What does it mean to turn any set of notes into a powerful database?

You must be fun at parties. Complaining about everything, but not even bother to read the damn manual... It's explained on the third sentence on that site. Ok, to be fair, there is a big picture between those parts, and you have to follow some links for more details.

> And why set of notes?

Just curious, do you even know Obsidian? Have you ever used it? You read like someone who has no clue about this software, jump right in the middle of the manual, and then complain that you missed the tutorial.

Obsidian is a markdown-editor with knowledge base. Notes are its lifeline, and they have since nearly the beginning the option to put metadata into each note in a special section (in yaml), basically the header of the note. This metadata are now called properties. Bases is a feature building up on this metadata, offering a database-like experience for viewing and editing them in a specialized UI. The database, is the vault, the folder+subfolder containing the notes. Each note is a row in that database.

This is all explained in the documentation, if you just would read it...

I use Obsidian every day, though I'm not a "poweruser" - I don't think I've ever installed an extension (presumably there'll be some default ones that are installed). Yes, I know Obsidian.

Yes, but the average Obsidian user may or may not know what a database is an why they should care. As an engineer I like precise language, but we should not forget that multiple audiences require multiple levels of explainations.

Otherwise it is a bit like saying "all monads are functors" when trying to make your reader care for investing time and energy into understanding the concept of monads. The problem there of course is that explaination is circular: without the reader knowing what a monad or a functor is they can't understand the explaination.

A good explaination gives you the technically correct slogan in the beginning (for the advanced readers) and then explains the words and what you can do with such a database and why you should care. Many explainations skip the last step and leave that part as an exercise to the reader.

> Yes, but the average Obsidian user may or may not know what a database

You can't constantly optimize your communication for the least educated recipient. Obsidian is full of technical, specialized terms. If you don't know a word, research it.

And database as a concept is common knowledge today. Everyone will have the word heard in news at least, most will have a rough understanding of its purpose. And people using obsidian are usually not the most uneducated, there have a certain level of technical expertise. Most will also know and understand the dataview-plugin.

Yes, of course you can optimize your communications. Humans tend to donthat constantly, this is just a matter (or: a mirror) of your priorities. If you care you optimize your communications and don't do if you won't. Whether you care about a particular target audience or you don't is up to you and nobody else.

However, it isn't exactly rocket science and if one developes such a feature it makes sense to test it with multiple realistic usecases anyways, so why not just use those to show the users how it can be used?

Unless ofc a feature is developed in mental isolation by a recluse with a niche interest in the code without thinking about the potential use at all. But then it is probably more like a extracagant hobby than a serious project that others should rely on.

Reply on news.ycombinator.com