Mindforger

Santema

Hi there, I just found https://www.mindforger.com haven't had time yet to completely check it out but it seems like something some of you will like. It has tags, live-markdown preview and best of all it's open source and available for Linux, macOS and Windows. There's more functions to.

ctietze

Great find! Thanks for sharing! We still feature open-source stuff here, so if anyone would write a review proper, and especially test its performance with huge archives, I'd happily add the app to the "Tools" page.

dvorka

Please don't review MindForger as Zettelkasten tool yet. I like the methodology, therefore I plan to add native Zettelkasten support to MindForger.

Christian, if you don't mind, I would be very interested in benchmark specification. What exactly do you mean by huge archive? I guess that you use per-Zettel file... can you please quantify number of files, their approximate size and sample content (one is just fine) so that I can generate benchmark data? In case you have a standard benchmark dataset, it would be even better!

I would be also interested in a baseline. MindForger is primarily Mardown editor/IDE, therefore I benchmarked several tools and editors, but didn't find faster competitor. Obviously it really depends on the structure, granularity and HW.

mediapathic

For what it's worth @dvorka , I pointed MindForger at my ZK of 1235 markdown files, and basic operations (by which I mean, me trying to figure out the software) were perfectly quick.

Sascha

@dvorka, Christian will give you a different number I guess. I think software should be able to handle at least 120000 files.

ctietze

@dvorka To get started, we have a collection of 10k Markdown files

For The Archive, I created 40-50k notes from books in the public domain to measure search performance. How fast does the app react when you type half a sentence? Does it stutter/block user input while searching? How fast do results appear for (1) all notes, (2) when you search for a single letter "a", (3) when you search for "and", etc.

Of course it doesn't make sense to share individual benchmark results across computers, only complete sets of comparisons, so you'd have to compare performance between apps on your own.

nickmilo22

No reviews until @dvorka says it's ready, but I just want to say this is wild and you have an interesting mind. Please do let this community know when you've given it your zettlekasten-ready blessing. I'll just say it's super snappy and seems as fast as any app out there.

sylvaticus

Brilliant! Kudos, @dvorka !

If I were using Windows still, or on Linux, I'd probably have already downloaded this! I'm still tempted. 😀

Question: Does MindForger work only with Markdown, or can it work with Multimarkdown as well?

Again, great job. Liked and subscribed. 🎉

dvorka

@ctietze said:
@dvorka To get started, we have a collection of 10k Markdown files

Thank you for nice dataset Christian - this is exactly what I wanted!

For The Archive, I created 40-50k notes from books in the public domain to measure search performance. How fast does the app react when you type half a sentence? Does it stutter/block user input while searching? How fast do results appear for (1) all notes, (2) when you search for a single letter "a", (3) when you search for "and", etc.

I measured boot, search by note name/full text/regexp and other basic use cases - looks good to me. As you pointed out, exact numbers depend on benchmark lab (machine) HW.

On of my goals is to make MindForger fast, therefore it trades speed for memory - notes are loaded to memory on boot, parsed, processed, indexed and made ready for use.

Thanks again!

dvorka

@nickmilo22 said:
... Please do let this community know when you've given it your zettlekasten-ready blessing.

I will!

dvorka

@sylvaticus said:
If I were using Windows still, or on Linux, I'd probably have already downloaded this! I'm still tempted. 😀

macOS?

Question: Does MindForger work only with Markdown, or can it work with Multimarkdown as well?

Yes, MindForger works with Markdown. I wrote own (fast) incremental parser and it uses cmark-gfm (GitHub) for Markdown rendering (most probably fastest and most robust parser available).

However, MindForger is designed to be extensible - it's just matter of demand and priorities.

Again, great job. Liked and subscribed. 🎉

Thanks - appreciate it!

dvorka

@Sascha said:
@dvorka, Christian will give you a different number I guess. I think software should be able to handle at least 120000 files.

@Sascha same file structure as in 10k repo (~5kB, ~3 sections)? Can I just augment 10k repo to 120k docs?

Sascha

@dvorka said:

@Sascha said:
@dvorka, Christian will give you a different number I guess. I think software should be able to handle at least 120000 files.

@Sascha same file structure as in 10k repo (~5kB, ~3 sections)? Can I just augment 10k repo to 120k docs?

I think you should ad approx. 10% Structure Zettel, complex lists and complex structured files in general.

1. Using the Zettelkasten Method some Zettel (Structure Zettel in general) will be complex, structured and longer.
2. Many people will not adhere to simple file structures. Some people will use this kind of software as a self organisation tool. Some people, will write longer text and use the software for general software production.

Therefore, you should add a good amount of files that are not as simple and short.