Logseq Times 2023-01-15: Logseq 0.8.16, AI Chat GPT 3 plugin, Discussion and Thoughts on Logseq
This week, we have lots of updates, with the main ones being the PDF drag and drop feature as well as @Bsunter Open AI / Chat GPT 3 Plugin update. Let's get to it.

With the end of the 2nd week of January with us, how are the New Year resolutions going? Are you already bored with writing Happy New Year in every piece of correspondence, sick of hearing New Year, New Me, well fear not, Logseq Times is here to whisk you aware for a few minutes so you can forget about all.
Oh wait, just like Manchester United's first goal being offside, with the New Year, I have decided to slightly re-order the post sections, New Year, New Me and all that.
This week, we have lots of updates, with the main ones being the PDF drag and drop feature as well as @Bsunter Open AI / Chat GPT 3 Plugin update. Let's get to it.
Event
With the Logseq Community giving so much to me, I want to give something back. So I present to you Unofficial Out of Office Hours Logseq Teatime.
Sunday 29 January - 16:00 Gult (GMT+4) For info
In this session, I am not envisaging anything too taxing, grab a Tea (or coffee, heck, even lunch) and chat with fellow Logseq users about Logseq, other software or life in general.
Sessions will not be recorded and will be off the cuff - no need to prepare anything unless you have something on your chest you want to share / ask.
Logseq News / Events
- "We’re planning to share more about our vision soon 🚀" message was recently published - will post any updates as soon as possible.
Releases
This week the devs provided us with 0.8.16, which includes some very nice updates, including:
Features
PDF highlight drag & drop
Fixed issues
- Capture open-url failure
- Query table not showing full property values and publishing failing on query tables
- Save code block shouldn't reset the cursor position
- Graph name with period didn't parse correctly, causing wrong block URLs
- Wrong behaviour when backspace with line-head text selection
- File deletion handling in Logseq Sync; Should trigger remote->local sync when idle
- Should not stop syncing when Android App enters background
- Header navigation of namespace pages in page preview popup
- Remove non-clickable space between links in the sidebar
Enhancement
- Bug report page, adding clipboard data inspector
- Check app awake from sleep
- New query inputs for advanced query - Document: Query Inputs
- Add a new option :ui/show-full-blocks? to show full blocks in references
- Prevent Ctrl+A from selecting the whole document
- Refine undo/redo button icons
- Make the weblink PDF filename to be human-readable for the annotations page
- Add delete option for blocks context menu
- Center content in search items
- Translations - pt-BR, pt-PT, fr and tr
You can see some of the main updates live in action here:
Logseq 0.8.16 just dropped!
— Logseq 🪵 (@logseq) January 12, 2023
These are some of our favorite changes:
• PDF highlight drag & drop
• Delete blocks with your mouse
• More inputs for advanced queries
Download the update in-app via Settings.
Want to see our favorites in action? See 👇
The latest release of Logseq is 0.8.15, which 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
After a few messages in the Disord concerning the documentation surrounding Logseq, I have started a new project, "Unofficial Logseq Docs". The current Official Logseq Documentation is good, but I feel that having it written within Logseq is a little confusing (at the moment) and can sometimes be overwhelming.
For this reason, I have taken the great work done by those before me and represented / amended / added to their work in what is hopefully easier to read and more digestible documentation. The aim is to:
- Focus on the "under the hood" features as opposed to terminology and basics (I may add these later on but I feel there are already quality videos / explanations dealing with this)
- Provide a brief explanation of the feature, how to use it, and an example
- Include a brief 1-minute video to show it in action.
The Unofficial Logseq Docs are still a work in process so please bear with me. New topics will be added every few days.
@DannyHatcher was jointed by @ToolsonTech and @OneStuttering to discuss Obsidian and Logseq. In the video, they each give their thoughts on both apps and discuss the main features, similarities and differences. Worth a watch:
@Bsunter takes AI in Logseq to the next level (yet again). He announced a big update to his Logseq Open AI / GPT 3 plugin. We now have a UI popup that "prompt templates" or writes freeform text as the prompt. You can ask the AI to perform any task on the current block, use a built-in command, or define your own templates. Check it out:
Really excited to announce a big update for the @logseq @openai #gpt3 plugin!
— Brian Sunter 🧠 (@Bsunter) January 13, 2023
A UI popup that lets you use "prompt templates" or write freeform text as the prompt.
You can ask the AI to perform any task on the current block, use a built-in command, or define your own templates pic.twitter.com/fpU2CcCGyg
Staying with Twitter for a moment, @TfTHacker's January deep dive into Logseq continues. He is more impressed with it each day and, in this thread, shares some highlights, along with some concerns.
1/12 My January deep dive into @logseq continues, & I am more impressed with it each day. In this report, I share some of those highlights, along with some concerns.
— TfT Hacker - Exploring Tools for Thought and PKM (@TfTHacker) January 13, 2023
I have a personal preference for Outliners, and if you are the same, Logseq is worth trying.
@Anthea Liles shows us a flow diagram and how they use Logseq, and other apps to manage content. Although the GitHub and GitBook points are not for me, it provide a useful visual guide of how you can go about storing content you come across.

@rrampr has a few questions about workflows. They currently use the journal page with links to pages being either important Proper Nouns (like [[Person]]) or Concept Nouns like ([[Atomic Number]] or [[Zettelkasten]] or [[Central Limit Theorem]]. They have also seen variants where pages are declarative or imperative phrases like [[Do your own Thinking]] or [[Spaced Repetition Systems make memory a choice]].
They ask:
- When is it appropriate to use block-level properties and page-level properties? I don't use them often and just rely on page references.
- When do you all find yourself using block references over page references?
- Should every top-level block in the journal page be linked to something so you can find yourself back?
- The biggest struggle I face right now is that I mix my personal learning/ "thinking" notes (that don't have actionability and are just centred around learning and personal enrichment and linking ideas) and notes that are for organizing my personal life (like logging visits to the dentist/doctor or notes related to a project or fitness trackers) in the same graph.
They find it hard to find the latter pages since they are all over the place. Linking seems great for learning as are on a timeline but not too great for the latter type of "organizational notes".
What are some workflows you use to make sure you find organizational notes quickly? Would love to hear your thoughts.
For those interested in local encryption after the feature was removed in Logseq, check this Reddit post on "How to go about encrypting your data locally". In essence, you can use something like Cryptomator, VeraCrypt or LUKS.
A comment I saw online about Personal Knowledge Management vs Personal Life Management made me think about the content we store in our Logseq, for example:
- Should we be storing birthdays, reminders, and recurring tasks which are more on the life aspect as opposed to the knowledge side of things in Logseq / our PKM?
- Do we take action on the great notes we are taking or is it more a case of feeling good about writing them but not bringing them forward? A bit like thinking of going to the gym, but not actually going there.
- Have we now gone to the opposite spectrum and feel the necessity to store every detail in our PKM instead of taking a step back and only including the important facts? A bit like transcribing every single word said in a meeting vs only noting the key aspects so that you can make sense of things as opposed to being overwhelmed by 10 pages of notes.
Hopefully someone else also thinks the above is interesting.
Want a way to quickly create text from the CLI to the journal page in Logseq? You can try out "dly - your daily note from the command line".
From the Readme: This small utility was built because I needed a quick way to add one-line entries to my daily notes from the command line. These would usually be some clever thoughts I do not want to forget, TODOs, things to buy, ...
I do not want to start Logseq just for that (even though I mapped it to shortcuts) and so dly
was born.
A question that often arises is how to remove certain menu items from the left-hand sidebar. Whilst there is no on / off toggle available in Logseq, you can hide unwanted options with CSS.
You can add the following to your custom.css to hide the option in the sidebar by
- Open your custom.css file (search custom.css and open)
- Add the following line
.texttext { display: none; }
- Change
texttext
with any of the following:
journals-nav
to remove the Journal optionflashcards-nav
to remove the Flashcards optiongraph-view-nav
to remove the Graph View optionall-pages-nav
to remove the All Pages option
@drawingthesun gives us some thoughts on a modular / tabular view of Logseq
We are all conditioned to the tab system, we click through tabs without thinking, it's like driving a car, we are people of the internet and tabs are a core component of that.
It's to such a degree now that many of us can navigate a stupid number of tabs with seemingly little friction, without realising it our brains have been moulded by this convention and we just don't think about tabs but have no trouble using them. (we only see through this and think about tabs if we ever go near a non perfect tab implementation)
I feel like Logseq would do very well to emulate the tab experience we get from a web browser. I use sidebars, split view sometimes, it's useful on occasion and I would love more development into that area.
However I would vote for putting more urgent development now onto making tabs a core feature of Logseq. I use the plugin and it makes such a big positive difference, but as you have pointed out there are issues, such as the tabs not being true independent views/instances - Opening tabs reloads the page, resetting the view to new each time.
It would be nice to have a native core tab system, for example, being able to drag a tab from one Logseq window to another would be super cool.
@TfTHacker has a cool trick. In Logseq you can assign properties to a page or block. In this case, I am assigning an "icon" property to the page. You do this by:
In the first block of the page, type "icon::

For those interested in Plugins on Mobile:
- one of the biggest issues with plugins on mobile (as well as on Desktop) is their missing UI consistency. Ideally, in a near future, plugins use a component system offered by Logseq, and those components should be out-of-the-box mobile optimized. We’re on it! Source
Logseq Feature Requests
You can check out the full list of Feature Requests here.
- Ordered List Items as Blocks - Could ordered list items be made into separate blocks? The way LogSeq currently handles ordered lists is an underrated but critical problem for note-taking and networked thinking. Beyond a CSS decoration, ordered lists have a semantic significance that makes visualizing sequences, listing and sections possible.
- Can I separate the pdf reading window and the note window? I have two screens and I want one for pdf reading and annotation and another for note-taking. Is it possible to separate the two windows?
- Distribution through Microsoft Store - would it be possible to distribute Logseq through Microsoft Store? I would like to use Logseq on my work laptop, but for security reasons we are only allowed to install apps through the Microsoft Store.
- Introduce UI elements to select spellchecker - It’s possible to add another language for spellcheck by editing C:\Users\UsErNaMe\AppData\Roaming\Logseq\Preferences (Windows example). If you do this both English and German are spell checked, even if it’s written on the same page. Example, added German: "spellcheck":{"dictionaries":["en-GB","de"]
- Publish whiteboard - I think it’s more like a bug fix since the published whiteboard is blank. But if it’s considered a feature, I guess here’s a feature request for it.
Logseq Plugins
- Doc View Exporter is updated to better adapt Logseq v0.8.15+. I also remade a video recording showing how to print a perfect PDF out of it. https://github.com/sethyuan/logseq-plugin-doc
- Logseq Plugin Leetcode - A plugin that can fetch leetcode problem in your logseq.It provides two commands: ltc all problems and ltc fetch problem.
- Logseq Grammarly Plugin - A plugin to toggle Grammarly inside LogSeq. Enable/disable Grammarly suggestions by clicking the green bell icon.
- @Luhmann offers a plugin bounty for Letterboxd integration. I have been using both Letterboxd and Logseq to track films, and I'm wondering if it would be possible to use the letterboxed API to make this work better. IMDB could work as well. I like Letterboxd better as an app, but the advantage of IMDB is that it also lists TV shows. Perhaps it could pull data from both?
Until Next Time
I hope you found this blog post helpful. If you have any comments or questions, please do just let me know.
Thanks again for reading.
Comments ()