⚡️ Ideas
Rahul Arora Rahul Arora Mar 27, 2023

LaTeX support, both inline and full-width

Explain the problem as you see it

I often take notes about technical content, including mathematical formulas

Why is this a problem for you?

This is one of the two things that is still keeping me on LogSeq for some of my use cases. The other thing is the lack of spaced repetition flashcards

Suggest a solution

Implement LaTeX with $...$ syntax for inline and $$...$$ syntax for multiline

⁨15⁩ ⁨Comments⁩

Yes. I agree. For the developers, this is a similar post to Steven Durham. I hope the votes would add up.

While I do understand Tana seems to like WYSIWYG style, some implementations could be similar to Ed, as shown in the photo below:
Screenshot 2023-04-06 at 4.41.55 PM.png

It can be seen that as a user, you can use latex form or a click the appropriate layout. so depending on their latex skill level, you can use either the former or latter, or even both.

The alternative one is similar to Craft Docs as shown below:
Screenshot 2023-04-06 at 4.45.06 PM.png

But I think Ed has a better design.

However, the missing element in both of these options is an in-line equation.
This is obviously solved in latex through $...$ syntax.

In the personal knowledge management community, I believe that a lot of people major in Math, Statistics, Physics, Chemistry, Engineering, Computer Science, Software development. Even Social Science makes use of mathematical and statistical tools. So Latex Support is vital for a note-taking tool.

Agreed. So important for any STEM field. Tana's been an absolute dream to use compared to other PKM's for me as a student, but not being able to use equations has me looking for alternatives that don't exist.

Lack of maths is the main reason we are not considering Tana in my group (physics research) so addition of this would be awesome. The syntax $..$ for inline and $$..$$ for display would be the best as it taps into muscle memory for anyone writing latex, see Logseq for a good implementation in an outliner and Obsidian for long-form writing.

For tana to compete in an extraordinarily large market, there are some simple things that need careful consideration. There is lots of complexity and flexibility but this (by definition) appeals to a smaller cohort of users when compared to basic 'quality of life' capabilities. My two needs that allow me to abandon other tools are:

Image flexibility and configuration
Some form of latex for mathematical functions
Every other tool I have looked at cover these to varying degrees. It will be a deal breaker at some point. The AI integration is strong enough - surely it's time to think a little more about things that (I assume) a large number of users will benefit from.

thanks

It would be great if Tana could implement this feature. I have been using Tana for managing my research projects, including experiments, ideas, and paper organization, and it works great. However, without the support of math typesetting, which is very important in my (and I believe most of the STEM majors’) workflow, I am not confident to fully switch to Tana from Logseq.

I hope this feature can get more attention from the Tana team.

Thanks.

In reply to M Matt M

But I am sure that people using these kind of tools mainly do it for projects and notes that are technical. It is beyond me how this basic feature is not addressed ASAP: this has way higher impact than any AI-related features could ever have on the typical outline, link it all users. This is especially true for Tana, which given its high entry barrier, is most likely used by tech people who are already used to working with Objects and Classes (aka supertags) in their education or jobs. [This is hence relevant for a well implemented code block, which is currently unusable].

As seen in many other PKMS, people want primarily a good base product. All the AI can wait, till the base product is not fully functional for base usage. Tana is not for writers, but it is not for techies either, without code and equations implemented. The only reason people are hooked is that the underlying concept and ideas are so good that the potential is huge. The potential, though, is not in AI, is in your node-supertag system: let it explode, then think about AI.

[From an AI dev]

This is so true - I'd really love to think the development team (who are doing a great job) are listening to their users. I wrote on slack about the kano model and how a lack of basic features leads to frustration. Whereas Exciters may impress but do not create frustration when they are missing.

Basic Features: latex, offline operation
Exciter features: AI ... Generative AI ... intrusive AI

I'd never switch apps for a novel AI feature .. I've left many that lack basic features :(