Logseq Times 2023-04-29: Logseq 0.9.4, Logseq OpenAI / ChatGPT, Sync, Super-tags, Corporate Usage and more
A look at the latest Logseq news, updates and thoughts during the past week. Logseq 0.9.4, Logseq OpenAI / ChatGPT, Sync, Supertags, Corporate Usage, Feature Requests and Plugins

Hi Everyone, with a third of the year gone already, I figured it was time to revamp the Logseq Times post a little to make it easier to read and digest. To achieve this, I have amended the "Logseq in the Wild" section by focusing on the most prominent findings of the week and giving you my take and commentary on the post.
What this basically means is that we are carrying on as normal but with some more commentary from me.
I hope you enjoy this change and get more out of these posts.
Logseq News
- Nothing much to report on from the Team this week. They seem busy working away on something special (more details below).
Releases
Logseq 0.9.4 was released this week and included several bug fixes and other improvements.
Features
- Element locking for Whiteboards
- Shape conversion for Whiteboards
Enhancement
- Add :remove-block-children? query option for advanced queries
- Plugin API: restore editing cursor state only for the input element in viewport
- I18n: German translation
Fixes
- Whiteboards: export grouped elements and zoom to fit selection shortcut
- Whiteboards: portal height calculation bug
- UI elements shake caused by lazy loading
- Mobile app crashes while trying to swipe the block that has any collapsed block
- No block collapse control on collapsed query block for mobile apps
- Two query table fixes - Comma-separated properties display and sorting by page for block queries
- Image renaming fails when resizing pdf highlight area on Windows
- Logseq Sync fails on non-ASCII filenames for mobile platform
The latest release of Logseq can be downloaded here. For Android (Download the latest .apk) and for IOS, you can download or update Logseq from the AppStore.
Logseq In The Wild
In-Built AI?
The week's main topic of discussion must be the work being done to potentially bring Open AI / ChaptGPT to Logseq. From the below image, we can reasonably hypothesise that:
- There will be a right sidebar for interaction with the OpenAI / ChatGPT
- Conversations will be saved in a new folder called
/chats
- There will be a copy conversation feature - I assume so that this can be pasted into your Logseq
- Some editing features and conversation options, possibility to keep chatting with the bot.

What I am not too sure about is the following:
- What the protocol for the above will be? Will this be a separate add-on so that users can choose to "install" or not? Will it be packaged into Logseq and will users have an "on/off" switch?
- If allowed, what data gets transmitted to OpenAI.
- Whether this will run through an API or packaged like Notion AI.
Very exciting times ahead, indeed. I am sure that the devs have considered Logseq's "Privacy first" motto when building the above so that it satisfies both the users who want AI capability and those who don't. Only time will tell, though.
I think this needs a bit of communication from the Team to confirm their intentions and stem any potential misinformation spreading.
How to Keep Your config.edn Updated
John_Doe provides us with an excellent solution to a pet peeve of mine, the config.edn
file. For whatever technical reason, when changes and new implementations are done on the Logseq config.edn file by the team, the changes and updates will not be brought to your file when updating the app (I guess it is not to overwrite personal tweaks you have made).
Anyway, this solution is a good way to keep up-to-date:
- Save default
config.edn
from latest Logseq repo
logseq/config.edn at master · logseq/logseq · GitHub 2
to some local file (in my caseconfig-origin.edn
) - Open your
config.edn
in Visual Studio Code (VSC) - Hit
Ctrl+P
and start typingCompare
- Choose
Compare Active File With ...
- Choose file you just saved (in my case
config-origin.edn
) - Now you can see the changes between you current
config.edn
and the latest one. You can copy past command from the latest file to your config.
A bit manual, but a start. Ideally, Logseq would compare both versions automatically and alert you of any potential changes.
Replicate an image from the Website
When the website update took place, a few images were used to display the capabilities of Logseq. The main one being the following:

Numerous users are now asking, "How do I create this". In short, the answer is with a query and some (or a lot) of CSS tweaks. I can help you out on the query portion and get a table of results with images and text, but I am afraid I cannot help with the CSS that has been applied.
Potentially way to use Logseq on Stringent Corporate Policy Computers
A user asked, "Need the ability to use Logseq via a web browser and from work where not allowed to install apps - is this possible?"
Yes, this should be possible. You should be able to go to https://demo.logseq.com/ in any web browser and select the location of your notes. Hopefully, you can use OneDrive / Google Drive etc. and can then sync between work and home computers.
However, a user pointed out, "keep in mind to Check your IT policy. A user's workplace had a massive no-no policy of using unchecked websites to store data locally". The best way would be (as suggested by Bas) "see if you can find a couple of people to join in to get it through the approval process. It's very tedious but often the only way."
From the Logseq Team, "We're aware of the need to offer either a proper web version or a portable version that can run off a USB stick. We're looking into ways to integrate our sync service in the web version down the line." Not sure this will get around the issue fully in case there are USB limits, but it could be a start.
Lazy Loading
A few users this week have stated their discomfort when dealing with long pages. For example, Soleone says, "Is there a way to "pre-load" all blocks in a page? It's always using pagination and constantly loading the rest of the page as I scroll. I have a huge page with hundreds of entries and every time I visit that page and I want to go to the bottom of it I have to keep scrolling and loading. Something like the ability to eager load everything right away would be appreciated. It currently makes working with this important page very painful"
Apart from the customary workaround of "splitting your page up into multiple pages", or use the "TOC Plugin". There currently does not seem to be a way around this limitation.
I am hoping that in the future, the database version of Logseq will give us the freedom of long pages and quicker performance. Testing long pages on other outliner apps based on a database does not appear to impact performance, so hopefully Logseq can do something similar.
I appreciate both sides of the argument. I work with Excel, which supposedly can have 1,000,000+ rows of data. Anything above a few tens of thousands and performance gets painful.
Sync Complications
yngwi, points out that "sync is rather dangerous in combination with the quick share feature on mobile.
Imagine you worked all day on the desktop writing the daily journal. After work, you relax on your couch, find something interesting on the internet on your mobile, tap on share to Logseq (that you haven't synchronized yet today) and see the shared quick capture added to an empty journal (you haven't done anything so far today on mobile) which then gets synced back, erasing your daily journal from the remote version. This already happened twice to me.
Fortunately, I had Logseq closed on the desktop so I could just copy/restore the original version, but if I had left it open, it would have synced on the desktop as well.
This is a very good point and something that I can imagine taking place. There are so many scenarios where the above could be replicated. Yes, you can possibly restore your workings from versioning, but there should be a more elegant solution. I am unsure of the best solution to this, but maybe it would be best for quick capture to go to a separate page?
Supertags in Logseq?
Another fantastic post by Blogbourri, who mentioned that achieving supertag-like functionality in Logseq is possible by using block properties smartly.
With some specific tags, block properties and query vocabulary, you should be able to create a dynamic table based on your requirements.
The full post is here, and it is worth taking the time to go through it if you are interested in this aspect.

Things to Consider When Using Logseq in Enterprise
A great video by Dario / OneStuttering and Bas / ToolsonTech concerning using Logseq in an enterprise setting. Video covers:
- Introduction
- Logseq's open-source licensing model
- Is it feasible for Logseq to be forked?
- Bas talks about the workshop that he is running
- Sync and collaboration in Enterprise - Using GitHub rather than Logseq Sync?
- Using Logseq in a team environment and discussing GDPR environment
- Discussion around other systems e.g. Microsoft Loop
- Bas' thoughts on search in Logseq
- How Bas merges notes with a plugin
- Bas' approach to work and content creation
- End-to-end encryption and security concerns
- Discussion on using plugins
- How Bas built his website and workflows
- Can you use 'export graph' to share meeting notes?
- More details about Bas' workshop
LogseqMastey Course by Dario
Dario / OneStuttering is the mastermind behind the Logseq Mastery Course. A course that will not only take you to the next levels within Logseq, but also give you a different way of thinking that you can apply in any setting.
The course was previously only priced in USD, but Dario now offers pricing parity options. So "Do you reside in a country with lower purchasing power than the US dollar? Logseq Mastery now has a parity pricing option for the full bundle."
All you need to do is visit Logseq Mastery and fill in a form. This is a great touch and more course creators should consider this. What is also great is that it is a one-time purchase for lifetime access. No recurring subscriptions.
P.S. Link is not an affiliated link.
justinsc1984 posted this image which I thought was a very clever way of showing their process.
Using various tools to capture information and then organise it in a select few apps (I am not sure of OneNote in this section). Having organised the notes in a few key places, these can be distilled in Logseq to generate ideas and content outlines. To express, although Justin has shown a few apps, I guess any app where you can publish your content could be shown.

Logseq Feature Requests
You can check out the full list of Feature Requests here.
- Deep and Native Integration between Logseq and LLMs (such as ChatGPT) - I probably don’t need to stress how powerful LLM can be in boosting productivity with knowledge work. I am aware there is a plugin that allows you to summarize content with LLM (ChatGPT) already, but I think there should be more to that. Here are two features that I imagine would make Logseq a killer app (if not already) for knowledge management:
- Logseq mobile sync in background - Logseq mobile app could have sync support (“paid” feature) in background. When opening the mobile app I have to wait for sync before writing to avoid conflicts.
- Dash (—) when typing two hyphens (–) - would it be possible to convert the two hyphens in one dash? I use this feature in Notion a lot. For me, Logseq is also good for writing small texts.
- Do not log time shorter than <specified duration threshold> - Please add a setting that sets the minimum amount of time required for an action to be logged, so we wouldn’t have so many “1s” “5s” in logbooks when we quickly toggle tasks to DONE. They pollute the files and the only way to delete them is via an external editor, making it a pain point.
- A Better Graph View: CSS Style Customization and Clickable Nodes on Mobile Devices - I’m a huge fan of Logseq’s graph view, and it’s one of the reasons I was drawn to the software. However, my biggest pain points with the graph view are the lack of styling customizations and the fact that the graph doesn’t work the same on mobile devices as it does on desktop.
- Allow to reference blocks in three ways: by ID, name, namespace - Idea is to have block names relative to containing page (see Block reference to allow ((page-name#block-ref-name)) not just UUID) or other parent blocks.
Parent block can be referenced by((mypage/my block parent))
.
Child block can be referenced by((mypage/my block parent/child))
. - Toggle on/off linked references for Daily journal and other created pages - We have linked references for every page within Logseq, sometimes these reference lists get very long. This in turn affects the performance of the application as well as making it more cumbersome to scroll through continuous pages like the daily journal section.
- Whiteboard Area Highlight - I should be able to toggle to see the area highlights on an off in my whiteboard.
- All pages filtering - Can we have logseq remember what your filtering preference is when you click back to All pages? Say you have clicked to sort your pages. If you go into a page and then go back to all pages it forgets and you have to keep filtering every time you go back.
- Search in current file on mobile client - menu item for search in the current file on mobile client
- Whiteboard inspirations from Xournal++ and Onenote
Logseq Plugins
- Logseq-plugin-excalidraw is a powerful and versatile plugin designed to seamlessly integrate Excalidraw into Logseq. With this plugin, you can effortlessly create and edit Excalidraw drawings within your Logseq.

- logseq-plugin-extra-editing-commands - plugin that provides additional editing functionality. NOTE: This plugin does not set any default key bindings. You first need to adjust the settings to your liking.
- Logseq Copilot v1.11.0 now supports the shortcut, the default is
ctrl+shift+U
on Win/Linux andCommand+shift+U
on Mac. Now is available in Chrome and Firefox, the version for Edge is still pending review. Shortcuts can customize via browser built-in functions for now. - ChatGPT Plugin V1.2.1 Released Whats New?: 1. Generate Flashcards Command - internally uses hidden prompt to generate good quality flashcards. 2. Outline output support 3. bug fixes
- Text Wrapper has updated the default configuration to include more buttons and more features by default. If you want to use the default configuration, please open the plugin configuration file after upgrading the plugin and clear the definition of the rules section (backup first).
Until Next Time
I hope you found this blog post helpful. If you have any comments or questions, please let me know.
Thanks again for reading.
Comments ()