Post @post@qoto.org

@jeffmarkel @pluralistic

It's not based on blockchain, same situation of IPFS. Look at this other part of the thread:

https://qoto.org/@post/109965516109644382

Anyway I believed in the Fediverse before Mastodon was a thing but now I admit it's a total disaster and I would like that others were honest enough to recognize the fundamental problem.

I don't want to support anymore the mafia scheme by some large Fediberse instances and the silence of the other ones.

@rieger_san@mastodontech.de @pluralistic

It should be something totally optional and not pushed in the face of users like that. Typical of projects that have to do with cryptocurrencies.

But Nostr approach using relays is interesting.

@rieger_san@mastodontech.de @pluralistic

What scares me about Nostr is that despite it doesn't seem based on a blockchain, it is developed by cryptocurrencies enthusiasts and I still have to figure out what's the relation between Nostr and cryptocurrencies. I installed Nostros client for Android and it pushes this "Zap" thing that looks based on cryptocurrences, on something called "Satoshi" and other things I don't even want to see.

@pluralistic

The #Fediverse has its own set of problems, fragmentation in particular, lead by instances blocking entire other instances without their users knowing the consequences.

For example I'm on qoto.org where I can speak freely but it is blocked by other instances with runarounds, so there is a portion of the Fediverse that will never see my posts.

In theory if you have to choose an instance and you have a list of people you intend to follow, it is not guaranteed that an instance that allows you to do that exist. If you create your own, it could still be blocked by other instances after a while for ideological reasons or even just because you pissed off who happens to be the childish admin of an instance. Their users are trapped in those instances with no idea that other portions of the Fediverse exist.

This undermines the notion of "joining the Fediverse": which portion are you actually joining? How will the situation change as you use it? How do you really monitor and control your experience? How can you enjoy a "free" platform if you have to think about politics between instances that often boil down to childish skirmishes?

-----

What do you think instead of the new approach by #Nostr based on relays? https://nostr.com/

@ellane @darren @EpiphanicSynchronicity

Markdown files and folders has always been my approach and I was convinced to adopt Logseq because I felt it didn't put at risk my workflow but gracefully improve it. Or at least I feel in control.

As I said Obsidian and Logseq are not much different on that aspect, it's just that Logseq needs to structure Markdown files as indented lists.

@darren

If I say "Obsidian users" I mean the ones I know on average. Why would you think I am referring to you? I didn't even know you use Obsidian...

I have discussed many times with that other person about Obsidian, Logseq and FOSS and my point has always been the same but argued again and again.

This started with an attack by the Obsidian CEO against Logseq out of nothing, followed by corporate blabbering and just a few weeks earlier there was a very misleading announcement about Obsidian new secretly developed .canvas format being made "open source".

So sorry, I'm a little fed up with the Obsidian topic.

@darren @ellane @EpiphanicSynchronicity

No, mine was a genuine reply and since I follow her I had elements to think she doesn't know about this.

@darren @EpiphanicSynchronicity

For context, here we are talking about two apps that both store data as Markdown files and they can even be used together to some degree.

The main difference is that Logseq is an outliner, so it mainly stores Markdown indented lists.

A Logseq user controls how much extra syntax to adopt just like does an Obsidian user with plugins like Dataview. So my argument is that there is not much difference in practice.

Additionally, I pointed out that a big part of Logseq are queries and you can use Datalog, that is a query language just like SQL but more expressive, used for example in Machine Learning.

Obsidian users known about Markdown but miss that it is not the only standard that can be adopted. Retrieving notes is a key feature in any PKM system and so it is valuable not to reinvent the wheel. Logseq using Datalog (both internally and exposed to the user) is a major selling point. This is just an example.

The big difference instead is Logseq being FOSS for the reasons I stated.

@ellane @EpiphanicSynchronicity

Congratulations, you have rediscovered as a user one of the principles of Unix philosophy, "Everything is a file".

FYI Unix is the family of operating systems where Linux and MacOS comes from.

The success of Unix is related to storing everything as plain text files and use a composable set of commands to manipulate them.

#EverythingIsAFile

@EpiphanicSynchronicity

What if there are no other apps that meet your needs?

What if your productivity depends on a very good plugin for Obsidian and the same experience can't be reproduced with anything?

Also what alternatives to Obsidian do you currently have?

@EpiphanicSynchronicity

You are replying repeating what Obsidian team says without actually reading what I said... this is exhausing but for the last time:

If I actually care about a piece of software I can improve it by myself or pay someone to do the work.

This is the key piece you keep skipping: in this world when you want someone to do some work for you you pay them.

Would you buy a house knowing that can be repaired only by the company that sold it to you? Would you risk see that company disappear and you having to abandon your house?

Why would it be different with software?

FOSS projects even have the advantage that fixes and improvements can be made once for all users, present and future, so you will have more people in the same situation as you willing to contribute or support (a portion of) the development.

@ellane @EpiphanicSynchronicity

#Logseq is #FOSS, that means:

- it's not only the product of a company but also a tool that belongs to all the users.
- if the company give up on Logseq the development can be continued by others even by re-organizing a new team of developers eventually financially supported.
- it can only "die" if there are better alternatives and the users move to them. As soon as it is relevant, it won't die.
- I will always be able to modify it to fix bugs or improve it by myself and even if I have not the skills, I can pay someone to do the work.
- tools to export data from it can benefit from knowing exactly how it works internally.

Additionally, Logseq:

- has HTTP/JSON APIs to power external tools including the ones to export data.
- it uses Datalog for queries that is a very powerful query language, much better than SQL and very popular in some fields.
- it's written in ClojureScript, a Functional Programming language, that makes easy to take a specific portion of the code and use it in other tools, again to export data to other formats without writing everything from scratch (there is already a CLI tool called logseq-query that let you query your graph without Logseq actually running).
- you can use HTML in it (as mentioned in Markdown specification) but also Hiccup that is a Clojure-friendly way to write HTML/CSS, basically allowing the user to remodel the application by adding buttons and even mini-apps in blocks because it can even evaluate ClojureScript.

Now, Obsidian is supposed to be just a Markdown advanced editor but its plugins need special syntax like Dataview's, that is not standard in any way, to provide the same functionalities shipped by Logseq by default and involving more standard and enstablished technologies.

Logseq on the other hand is the convergence of outliners like Org-Mode, a graph database to power a flexible and performant data structure, Datalog query language, everything exposed as HTTP API for max interoperability and to run it over a network and this unique approach of the user remodelling the UI with HTML/CSS/JS, blurring the line between users and developers and between an app and a framework.

Logseq was born as a personal project to provide a modern UI to OrgMode, that 20 years later its first release is still very popular, gets new releases and has a huge ecosystem.

Thinking in terms of "will it still be around?" is something that belongs to the realm of proprietary products and services. Logseq is a piece of that huge amount of software that belongs to humanity known as FOSS. And this is also the reason we are discussing this on Mastodon and not on Twitter.