After a somewhat crazy last few months, I am finally properly back online and slowly getting back to normality, which gives me great pleasure to be able to post on Logseq Times once more. My last post was about 4 months ago, which is a little shameful, but sometimes it is important to get off the merry-go-round that is life before you start missing the trees from the forest.
Anyway, log metaphors aside (see what I did there), there has been so much going on in the world of Logseq of late, so why don't we jump right in?
Logseq 0.9.20 Release
This release includes a couple of useful updates like a new inbuilt Handbook, smartt-merge for this using Logseq Sync, updated Electron app and more. You can download the desktop and Android app here.
- New handbook (click in the app to see it - info)
- Upgrade electron to v25
- Make Smart Merge the default option for Logseq Sync
- Add toggle option for PDF view: auto opening the highlight context menu from new text selection
- Support zoom via touchpad
- Update Spanish, Norwegian and French translation
- Fix #card should only be appended to the title
- Fix Android back button closing order
- Fix journal name judgement
- Refine drag and drop of assets. Fix not being about to save assets sometimes
- On Android, activity restarts when a Bluetooth keyboard gets connected
- Shortcuts editor can't edit disabled keymap
- Refine Logseq Sync login box tips
Logseq Properties Coming Along Nicely
A lot of work is going into the upcoming properties and database mode, and it is starting to show. I think this will push better structure and knowledge management in Logseq and I cannot wait to see it live:
Properties Best Practice?
Talking of properties, OP asks: "Best practice - double-bracket the properties or not?"
An interesting response:
I think there's no "best practice" for this, since both approaches are valid and have their own benefits.
When you don't add the brackets, it is just a string. When you do add brackets, it becomes a link, which will make this block/page appear in the linked references of both its property page, and the linked value page.
Personally, I use string-type values as a "fallback" when the value isn't an entity in my graph.
Also, for the tags property specifically, it doesn't matter if it has double square brackets or not, logseq will always understand as link-type values.
Bring the power of local LLMs to Logseq with (Ollama-Logseq)
This should probably be in the Plugin Section below, but I thought it was too important to miss.
OP posts "I was jealous that Obsidian had a plugin integrating with Ollama, So I decided to make one for Logseq myself. Ollama essentially allows you to play with local LLMs like: LLama 2, Orca Mini, vicuna and many more. Some of these LLMs are performing to levels close to chatGPT3.5 in some tasks and the fact that they are running locally is awesome"
My good internet buddy Dario recently posted a video on Anti Zettlekasten where he considers building a great Zettelkasten can be a lot of fun, but it can also be a great procrastination tactic. For most writing workflows, all you need is a good system of tags, and to trust the process as you start writing.
Check out the great video here
Sticking with Dario, he also posted a video on his writing process, where he uses apps like Tana + Logseq + Obsidian. In this video, he ponders, "A lot of people will tell you you have to choose one application. I don't think that is the best way to go about things. Each application has different strengths, and in this video I quickly go through my process of writing an article, leveraging the different strengths of the apps."
Which way is *better* for creating meaningful connections in a general sense
OP asks a very interesting question: let's say I want to write something about a movie in a journal.
- I watched [[Movie/ABCD]] today - point A - point B
- [[Movie]] - [[ABCD]] - point A - point B
- I watched [[ABCD]] today #Movie - point A - point B
Which way do you prefer? The most popular response seems to be doing:
Journal - I watched [[ABCD]] today [[ABCD]] - tags:: #movie - point a - point b The page [[movie]] would then list all movies
Which way do you prefer?
20 Logseq Features You Didn't Know Existed
Despite the quick clickbaity title, Alan posted a great video on some of the potentially more obscure Logseq features. Definitely worth checking out it you are just starting using Logseq or are a power user, there is certainly something for everyone.
Build a Second Brain in Logseq
Gregory at Face Dragons, posted a fantastic article on using Logseq to build your Second Brain.
There is more second-brain software than ever. You’re so spoilt for choice that you may wonder, “why should I build a second brain in Logseq?” Whether you still need convincing or just want to know how to create a second brain in Logseq, this article is for you.
The features that make Logseq a contender from all the second brains apps available are listed below. You’ll also learn how to set up your second brain and which templates, plugins, and practices are best to extend your mind with an effective PKM setup.
“Introducing ‘Views’: A Flexible Solution for Graph Organization and Content Separation in Logseq”
A very interesting post on different views in Logseq to allow for better information / knowledge management.
Using Logseq for the Past Year
OP has used Logseq for the past year and this seems to have gone down well. Although the graph looks pretty, I would love to know how Logseq changed / helped / assisted OP in levelling up.
"Ditching" Notion for Logseq for 30 Days
Mariana Vieira, a long-time Notion user and expert, recently posted a video on "ditching" Notion to use Logseq for a 30-day period.
There is of course quite a large disclaimer necessary, "I manage my entire business in Notion so of course I wasn’t going to switch business ops to Logseq for the sake of this video; this experiment was basically for my personal note taking and organization and I did learn a lot about how Logseq works."
A great video, nonetheless.
- Full House Templates 3.2 Big Update! Fuzzy search by name, type, and parent page, Open page with the template by holding «Shift» key, Hide items from list with «.», Restrict to View or Template only, Provide usage hint, Control cursor position or text selection for arguments, Introduced UI for inserting templates & views, Specially designed to be very productive with keyboard
- Show page date plugin - Display creation and modification dates on pages. (Displays dates of each page to the right of the page name.)
- Quickly PARA method plugin Updated! Inbox feature(If interrupted, a link can be saved to the inbox page. It is categorized by month.). "namespace query search" feature (Search for pages with the same name or explore related pages without worrying about the hierarchical structure and list them.). The ability to create a new page in the same hierarchy or add a sub page is provided.
- File Linker. It can dynamically replace a link's path. It allows you to store big files (like video) outside of Logseq.
- Tabbed Sidebar update: Moved tabs are now remembered and restored as necessary. With this feature implemented, layouts like this now can be persisted.
- Side Block plugin - Places bullet blocks on the left and right. If tag parent blocks, its descendant blocks are placed next to it. Normally, it can be applied as custom CSS, but we have made it a plugin to make it easier to use.
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.