Logseq Times 2023-03-12: Logseq 0.8.18, CLI feature, workflow and new plugins

My apologies for the lack of LogseqTimes posts over the last few weeks. The inlaws, some traveling and work all took a little too much out of me. Happy to say that things are returning to normal so new releases will be back on schedule.

Logseq News

• @ToolsonTech and @Ramses got together to do a @logseq deep dive. They aimed to talk about the future of knowledge work, how they manage the daily deluge of information, and how Logseq fits in the picture. The video is being edited and should be available soon here.
• The Logseq Github readme had a bit of a makeover and modernization - be sure to check it out here.
• @cldwalker provided some tool announcements for the Logseq power / technical users out there.  A new Github action and CLI that automates the export of a portion of a Logseq graph to RDF. A node library for scripting with Logseq's core. Release 1.2.168 provides support for node CLIs. See here for more. Github action and CLI that runs validations on a graph. Release 0.3.0 adds more validations and the ability to exclude certain validations
• I won't pretend to understand what any of the above means, but hopefully can be of use to some of you.
  

Releases

Logseq 0.8.18 was released and included the following:

Features

• Support cues for Cloze cards - Documentation
• Update dependencies (bump NodeJS from 16.x to 18.x LTS, bump Ubuntu image from 18.04 to 22.04)
• New contributing guidelines
• Fix: incorrect page title when hls__ is opened
• Updated German and Turkish translations
• Add repl setup to development document
• Fix incorrect version control description
• Enhance pasting from clipboard
• Add command for copying a page url
• Fix "Copy Image" menu item correctly copies images
• Add description to plugin settings headers

Fixed issues

• Massive bug fixes for whiteboards
• Command completion triggering too often
• Handle duplicated block ids
• Tag and property value autocompletion cases
• Disable image resize on mobile platform
• Cannot copy blank string
• Remove prefix(@) checking for links of org mode page
• Force disable the unloaded(error) plugin
• Fix marker layout of area highlight PDF ref in block content
• Fix Logseq Sync network issues; add FreeBSD dev support

Enhancement

• Experimental support of auto-completion UI for config.edn file editing
• Add support for block refs in properties to backlink to blocks
• Refactored copy/export
• Whiteboards: Auto pan canvas to pointer when out of view
• Handle invalid config.edn
• Shortcuts can be disabled with false now
• Enhance re-index performance
• Show git status when commit changes
• PDF viewer enhancements
• Make the modals position friendly for the auto-complete component
• Local HTTP Server now supports CORS for use from another web page or browser extension

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

Workflow for PDFs? I’ve been experimenting with PDF annotations inside of Logseq. I typically create a page, attach/upload a PDF, and then away I go. But then when I look at my graph I see a bunch of separate annotation pages that are automatically created. So I end up with two separate pages in my graph for the same PDF. Is that normal? How is everyone else working with PDFs?

A few responses to the above include: 1) Use Zotero, 2) Use Omnivore - does anyone else have any suggestions?

Change spellcheck language?  Is there a way to change the spell check language, or even better, to add a second dictionary? I fairly fluidly go back and forth between languages (English and French), so having both spellings accepted without all the underlines would be quite nice.

Locate the Preferences file (mine was located  here C:\Users\XXXXXX\AppData\Roaming\Logseq) and change:

{"spellcheck":{"dictionaries":["en-GB"],"dictionary":""}}

to (for example)

{"spellcheck":{"dictionaries":["en-GB", "fr"],"dictionary":""}}

Restart Logseq and you should be good to go.

Does Logseq query have a Count function? So I can count the total by adding up a certain value of a property?

Yes, with query table functions: https://docs.logseq.com/#/page/64076691-bea5-429c-a43c-dc9ea8e31e01. In short {{function (sum :amount)}} where "amount" is a property of the results of a query defined in the parent block.

How to download an uploaded asset?

After uploading (/upload) a PDF to a note, when I click on it later it brings up a small view of the PDF (in split screen mode, in the left half) so I can see it. But I cannot see any way to download the pdf from there. How do I do that (Apart from hunting in the assets folder)?

Remember that you can download pre-filled templates from https://www.logseqtemplates.com/

There are templates for:

• Books
• Projects
• Persons
• Tasks
• Journals

You can download and then tweak to suit your needs.

Some tips on using Logseq for time tracking:

How easily does Logseq handle a few big pages? I want to avoid using namespaces, so planning to make do with indents and blocks for some hierarchy. Any anecdotes from those who have big pages with a lot of text content?

qwqpwp suggests: when you have to work with long pages, 1. open this page 2. shift-click all the smaller sub-trees you might need into the sidebar (or cmd-click to tabs if you use the plugin), so you only work with the shorter bits to avoid lazy loading 3. store this page somewhere (favorites list, right sidebar and collapsed, contents page, doesn't matter) so you can quickly access it when the pre-opened nodes weren't enough for you - but you'll have to tolerate lazy loading for that

Hi, a beginner question here. I have a huge folder in my computer with my favourite books. I love how logseq permit annotate pdf in a very natural way. The problem is every time I "/upload asset", I'm creating a copy of every file in the logseq assets folder. So, my computer will end up having two huge folders. is there a solution for it?. (I'm using ubuntu). Thanks!

https://discord.com/channels/725182569297215569/735747000649252894/1082994916675825714 another way is don't "upload", just embed the asset to the original filepath via the existing markdown syntax (but it creates the problem of not being able to see it on another synced platform)

Querying Now and Later tasks?

The query should be: {{query (and (task NOW) (task LATER)) }}. Check out the guides written by Ramses here.

With the starting point of writing and taking notes in the journal, do folks usually try to keep their journal pages "clean" and orderly? Or is it better to braindump, and utilize it as a kind of scratchpad and then create pages that are better organized? I know this is going to be dependent on the individual but that's my point in asking.

A few responses: I mainly braindump in my journal. If I find to need something more open I move stuff to a separate page - but I leave a trace in the journal. Like, what's left afterwards is in the journal like "Read about topic a, these detail"

I use my journal page as a brain dump. all my notes go here in no particular order aside from parent/child necessary relationship

I'm also a uni student using Logseq for notes, here are my tips:

• If you take a lot of notes, make a separate graph for each class (unless you need to link between your notes for different classes, bc linking between graphs is a little funky)
• afaik there is no vanilla support for executing code blocks. Idk if there's an addon for it.
• The PDF export plugin doesn't work super well. Template 1 is the best option, Template 3 is entirely broken. I've been trying to figure out a good export solution with Pandoc, but haven't managed it yet.
• Making use of the tags and card features is great.
• Tagging a block that has child blocks with #card automatically makes it accessible as a flash card via the Cards tab. The block tagged with #card becomes the prompt, the child blocks become the answer.
• You can use the built-in excalidraw integration with /draw to make drawing diagrams really easy. If you want it to be saved as an image instead of an excalidraw canvas when you're done, right click and save as PNG, then replace the draw canvas with the PNG.
• Using the block embed feature to make the same To-Do list block accessible from multiple pages is extremely useful.
• Right click on a block and then click copy block embed, paste the embed into another block.

OH also it has inline LaTeX integration that renders as soon as you click off of the block, it's wonderful

You can type "my fun theorem is defined as $\sum_{k = 0}^{i} C_x C_{x+1}$" and the LaTeX will render nicely (I have it in a code block to prevent discord from trying to treat the TeX as markdown)

I'm using Logseq as my learning platform. The Logseq Anki Sync Plugin is great. On my right Sidebar I've placed all my subjects. You could use hierarchies with a / in the page name. Give yourself a couple of weeks to try out other systems. Remember: When you decide on a system you use it the whole time. If you already have about 100 Notes you have to be skilled to change your system. So give yourself a couple of weeks. This helped me a lot. Install all plugins you like and then look if you really need them.

Interested in a Plugin Bounty? Someone is after a Shazam Plugin - The plugin would need to show all shazammed songs based on the shazam history.

Does anyone have a good workflow for taking key points out of a longer series of notes and highlighting them in a way you can view with the longform notes collapsed in the tree? This would be especially useful on my journal page, where I would use this most.

How about something like this?

Interested in a book about Personal Knowledge Graphs. Well, look no further than here.

As knowledge graph enthusiasts and practitioners, we are happy to note the quantity and quality of literature on the topic. From research publications and books to blog posts, there is a growing body of knowledge on knowledge graphs.

What's missing from this picture however, is a book on PKGs. This is what we want to address. We are looking for thought leaders and innovators from all walks of life and backgrounds: practitioners, academics, developers, power users and beyond.

We want you to embark on a journey of exploration with us, to share your views, experience, theories and achievements. The book aims to cover a wide spectrum of concerns, from onboarding, use case and design, to theories, technologies and best practices. We find it's best to show, rather than tell. This is why the book itself will be published as a knowledge graph, although more "traditional" formats will be produced as well. 

Logseq Feature Requests

You can check out the full list of Feature Requests here.

• Open pdf in a separate window - I have two monitors and want to read pdf on one monitor and take note on the other. How could I achieve that? ty!
• Show full URL on hover Feature Requests - Although you can embed clickable links to web pages in your Logseq blocks, they’re not as easy to work with as links in a web page. When hovering the mouse pointer over a web link, it would be good if the URL was visible in a tooltip or in a floating status at the bottom of the window - just like a web browser. This would enable you to see where the link went before you clicked it. It would be good if right-clicking on a web link gave a context menu that included a “Copy URL” option. This would allow you to grab a URL without having to visit the page first. Both reading and copying a URL can currently be achieved by editing the block but these changes would make those tasks more convenient
• URL Scheme for searching text - I found Logseq Protocol for opening the graph to a specific page/block and for quick capture. Is there a URL scheme for searching text? something like logseq://graph/<graph name>?search=<search term>
• Block References Issues and Ideas for Improvements - I’m trying to leverage block references more in my notes, and I’m experimenting with a workflow where I write most of my notes in my daily notes, then create dedicated pages with references to blocks in my daily notes, but I’m running into some usability issues.
• Add pagename field to the list of table fields in simple query - When I do a simple query using a page-property I can select the pagename from the list when I want the query to show as a table.
• PDF pages should have their own section and priority in search - I often tend to webclip webcontent as PDF just to make use of this awesome annotation feature. Workflow is to treat these PDFs just as annotation container, which is then linked to my own notes.

Logseq Plugins

• Logseq OmniFocus Plugin - A simple plugin for sending tasks from Logseq to OmniFocus.
• awesome-content - Enhanced content blocks (tasks, quotes, flashcards, headers, queries, diagrams, etc...) plugin for Logseq including - Colored tasks statuses & priorities, #.kanban & #.grid columns (no plugin needed), Redesigned admonition blocks, Custom styled #quote, Custom styled Mermaid diagrams, Compact QUERY results header: settings, table toggler, Today queries as columns in journal, Redesigned calculator.
• Full House Templates - Super-configurable, JavaScript-based Logseq Templates for Humans focused on UX and simplicity. This is the missing power in your Logseq knowledge base.
• logseq-lilypond - This plugin has a simple function, that is, allow rendering lilypond in logseq!
• Introducing a tightly integrated chatGPT extension for Logseq. UI is not complete and will change. Several other goodies not complete as well. This is an extremely incomplete development preview.
• Logseq Copilot! Browser extension now includes some amazing new features: - Show the record you have been seen (based on URL) - Popup your history record on this page (based on URL) - QuickCapture when you select on a web page (note function coming in the next release) With these new features, you'll be able to browse and capture information more efficiently than ever before.
• logseq-query-builder-plugin is now in the marketplace. Thanks to the community for assisting me to have the plugin migrated into the marketplace.

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.